阅读说明：本技术 用于重建药物应答和疾病网络的方法和系统以及其用途 (Methods and systems for reconstructing drug response and disease networks and uses thereof ) 是由 B·D·阿西 G·A·希金斯 A·艾德 A·卡利宁 N·里马鲁恩 J·S·伯恩斯 于 2020-01-22 设计创作，主要内容包括：描述了包括重建药物特异性药物基因组学网络和其组成功能子网络的集成的、多尺度的、基于人工智能的系统的方法。所述系统使用染色质空间中的药物调制的空间接触的三维架构的功能拓扑的特征。药物药物基因组学网络的发现是通过以下进行的：借助于插补来选择候选SNP、使用机器学习和深度学习来确定所述SNP的预测因果关系、使用因果关系SNP来探测如通过染色体构象捕获分析确定的空间基因组、将由同一细胞以及组织特异性增强子控制的靶向基因组合以及基于全基因组关联研究的结果使用不同数据源和度量来重建药物基因组学网络。使用基于知识的分段方法来将所述药物基因组学网络解构为其组成功效和不良事件子网络,以用于应用于临床决策支持、药物再利用和硅中药物发现中。(Methods are described that include an integrated, multi-scale, artificial intelligence-based system that reconstructs drug-specific pharmacogenomic networks and their constituent functional sub-networks. The system uses features of the functional topology of a drug-modulated, spatially-contacted, three-dimensional architecture in chromatin space. The discovery of pharmacogenomic networks was performed by: selecting candidate SNPs by means of interpolation, using machine learning and deep learning to determine the predicted causal relationships of the SNPs, using causal SNPs to probe the spatial genome as determined by chromosome conformation capture analysis, combining targeted genes to be controlled by the same cell and tissue specific enhancers, and reconstructing pharmacogenomic networks using different data sources and metrics based on the results of genome-wide association studies. Knowledge-based segmentation methods are used to deconstruct the pharmacogenomic network into its component sub-networks of efficacy and adverse events for application in clinical decision support, drug reuse, and drug discovery in silico.)

1. A computer-implemented method for identifying a pharmacogenomic network of drugs, the method being performed by one or more processors programmed to perform the method, the method comprising:

obtaining, by one or more processors, a plurality of Single Nucleotide Polymorphisms (SNPs) associated with a drug response or an adverse event of a particular drug;

identifying, by the one or more processors, a pharmacogenomic network for the particular drug based on a gene set associated with the plurality of SNPs;

deconstructing, by the one or more processors, the pharmacogenomics network into a plurality of pharmacogenomics subnetworks using a topology of subnetwork types selected from two or more of: a chromatin remodeling subnetwork type, a drug efficacy subnetwork type, a drug adverse event subnetwork type, a pharmacokinetic enzyme and hormone subnetwork type, a systemic pharmacokinetic subnetwork type, or an immune system adverse event subnetwork type; and

providing, by the one or more processors, an indication of the pharmacogenomic network and an indication of the plurality of pharmacogenomic subnetworks for display.

2. The method of claim 1, wherein identifying a pharmacogenomic network for the particular drug based on a gene set associated with the plurality of SNPs comprises:

comparing, by the one or more processors, the plurality of SNPs to a SNP database using Topological Association Domain (TAD) boundaries within a chromosomal domain to identify additional SNPs that are linked to the plurality of SNPs, wherein the plurality of SNPs and the additional SNPs are included in a candidate set of variants;

performing, by the one or more processors, mapping of 3D spatial junctions using the set of candidate variants as probes within chromatin data to determine interconnections between target genes associated with a drug response or adverse event of the particular drug;

performing, by the one or more processors, pathway analysis on the target genes associated with the set of candidate variants to filter the target genes to identify a set of genes causally related to the particular drug; and

identifying, by the one or more processors, the pharmacogenomic network for the particular drug based on the identified gene set.

3. The method of claim 2, further comprising:

performing, by the one or more processors, a bioinformatic analysis on the pharmacogenomics network to ensure that the drug most significantly associated with the plurality of pharmacogenomics subnetworks is the particular drug.

4. The method of claim 3, wherein deconstructing the pharmacogenomic network into a plurality of pharmacogenomic subnetworks comprises:

organizing, by one or more processors, the gene set into functional subsets using iterative gene set optimization to identify a plurality of sub-networks of drug pharmacogenomics for the particular drug corresponding to each of the functional subsets.

5. The method of claim 4, wherein performing bioinformatic analysis on the pharmacogenomic network comprises:

verifying the pharmacogenomic network and the plurality of pharmacogenomic subnetworks of the particular drug by bioinformatics.

6. The method of claim 5, wherein validating the pharmacogenomic network and the plurality of pharmacogenomic subnetworks for the particular drug comprises:

comparing the pharmacogenomic network and the plurality of pharmacogenomic subnetworks of the particular drug to one or more of:

terms from gene ontology or drug database;

typical biological pathways in the cell or tissue in which the particular drug acts, or

A modulator of xenobiotic upstream in said cells or said tissue on which said specific drug acts.

7. The method of claim 2, wherein performing pathway analysis on target genes associated with a subset of intermediate candidate variants to filter the target genes comprises identifying, by the one or more processors, a candidate drug pharmacogenomics network gene set by identifying a subset of the target genes that form a statistically significant interconnected pathway expressed in a tissue associated with the particular drug.

8. The method of claim 2, further comprising:

analyzing each gene in the candidate drug pharmacogenomic network gene set according to at least one of: a function of the gene in the context of the particular drug, a set of mutations within the gene, or a pattern of expression of the gene relative to a neuroanatomical substrate, as defined by RNA expression data, functional imaging, or other integrative multi-scale data indicating where the particular drug is known to act;

analyzing other genes and functional genomic elements comprising long non-coding RNAs based on each gene in the candidate drug pharmacogenomic network gene set; and

adding or deleting genes to or from the set of candidate drug pharmacogenomic network genes based on the analysis.

9. The method of claim 8, wherein analyzing each gene in the set of candidate drug pharmacogenomic network genes for a pattern of expression of the gene relative to neuroanatomical substrates comprises:

comparing each gene in the candidate drug pharmacogenomic network gene set to a neural map indicative of the neuroanatomical substrate of the particular drug; and

filtering from the candidate drug pharmacogenomic network gene set genes that are not expressed in the same neuroanatomical region as the particular drug.

10. The method of any one of claims 1-9, further comprising:

determining, by the one or more processors, one or more spatial contacts comprising a TAD based on the plurality of Single Nucleotide Polymorphisms (SNPs) associated with a drug response or adverse event of the particular drug, wherein the one or more spatial contacts are differentially expanded and inhibited.

11. The method of any one of claims 1 to 10, wherein the pharmacogenomics network comprises one of more defined characteristics of functional spatial genomics, the more defined characteristics comprising at least one of:

mutations affecting enhancer-promoter pairs;

a promoter-promoter pair;

super enhancer-promoter pairs;

a regulatory RNA;

euchromatin or heterochromatin state;

a Topology Association Domain (TAD); or

A Lamina Association Domain (LAD).

12. The method of any one of claims 1-11, further comprising:

storing, by the one or more processors, the indication of the pharmacogenomic network for the particular drug and the indications of the plurality of pharmacogenomic subnetworks in a database as a reference set to compare and score input patient pharmacogenomic networks for the same particular drug to determine efficacy and adverse events in treating the condition or disease of the patient.

13. The method of any one of claims 1-12, further comprising:

identifying, by the one or more processors, a set of molecular pharmacodynamic targets within one of the plurality of pharmacogenomics subnetworks for a same or similar clinical indication as a specified drug.

14. The method of claim 13, wherein the set of molecular pharmacodynamic targets and the particular drug within one of the plurality of sub-networks of pharmacogenomics functions in the human brain, and a majority of members of the pathway of the molecular pharmacodynamic targets are expressed in the same region in the human brain as the particular drug.

15. The method of any one of claims 1-14, further comprising:

reusing drugs based on potential loci of pharmacogenomic enhancer SNPs associated with the plurality of pharmacogenomics subnetworks within the topology of the subnetwork type.

16. The method of any one of claims 1-15, further comprising:

determining complementarity between sub-network properties of two or more drugs exhibited by physiological mechanisms provides for the reuse of a combination of drugs to be tested as a therapeutic agent after a therapy that is superior to a single specific drug for a particular disease or disease state.

17. The method of claim 16, wherein the combination of reutilizing drugs comprises reutilizing valproic acid and ketamine (ketamine) as a combination therapy for neurological and neuropsychiatric disorders.

18. The method of any one of claims 1 to 17, wherein the specific drug is one of: valproic acid, ketamine, lithium, lamotrigine (lamotrigine), clozapine (clozapine) or warfarin (warfarin).

19. A computing device for identifying a pharmacogenomic network for a drug, the computing device comprising:

a communication network;

one or more processors; and

a non-transitory computer-readable memory coupled to the one or more processors and storing instructions thereon that, when executed by the one or more processors, cause the computing device to:

obtaining a plurality of Single Nucleotide Polymorphisms (SNPs) associated with a drug response or adverse event of a particular drug;

identifying a pharmacogenomic network for the particular drug based on a gene set associated with the plurality of SNPs;

deconstructing the pharmacogenomic network into a plurality of pharmacogenomic subnetworks using a topology of sub-network types selected from two or more of: a chromatin remodeling subnetwork type, a drug efficacy subnetwork type, a drug adverse event subnetwork type, a pharmacokinetic enzyme and hormone subnetwork type, a systemic pharmacokinetic subnetwork type, or an immune system adverse event subnetwork type; and is

Providing, by the communication network, an indication of the pharmacogenomic network and an indication of the plurality of pharmacogenomic subnetworks for display.

20. The computing device of claim 19, wherein to identify a pharmacogenomic network for the particular drug based on a gene set associated with the plurality of SNPs, the instructions further cause the computing device to:

comparing the plurality of SNPs to a SNP database using Topological Association Domain (TAD) boundaries within a chromosomal domain to identify additional SNPs that are linked to the plurality of SNPs, wherein the plurality of SNPs and the additional SNPs are included in a candidate set of variants;

performing mapping of 3D spatial junctions using the set of candidate variants as probes within chromatin data to determine interconnections between target genes associated with a drug response or adverse event of the particular drug;

performing pathway analysis on the target genes associated with the set of candidate variants to filter the target genes to identify a set of genes causally related to the particular drug; and is

Identifying the pharmacogenomic network for the particular drug based on the identified gene set.

21. A method for identifying a combination of drugs to be tested as a therapeutic agent for a particular disease, the method being performed by one or more processors programmed to perform the method, the method comprising:

obtaining, by one or more processors, a first plurality of Single Nucleotide Polymorphisms (SNPs) associated with a drug response or an adverse event of a first drug;

obtaining, by one or more processors, a second plurality of SNPS associated with a drug response or adverse event of a second drug;

identifying, by the one or more processors, a first drug pharmacogenomic network for the first drug based on a first gene set associated with the first plurality of SNPs;

identifying, by the one or more processors, a second drug pharmacogenomic network for the second drug based on a second gene set associated with the second plurality of SNPs;

identifying, by the one or more processors, a first drug response phenotype associated with the first gene set in the first pharmacogenomic network that is complementary to a second drug response phenotype associated with the second gene set in the second pharmacogenomic network; and

the to-be-tested combination of the first drug and the second drug is then utilized as a therapeutic agent for a particular disease.

22. The method of claim 21, wherein reusing the to-be-tested combination of the first drug and the second drug as the therapeutic agent for the particular disease comprises:

administering the first medication at a first time; and

administering the second medication at a second time later than the first time.

Technical Field

The technology described herein relates to the discovery of gene network contacts in chromatin space that define pharmacogenomic substrates for a particular drug, the identification of functionally distinct sets of genes within the pharmacogenomic network (referred to as subnetworks) of that drug, and the detection of regulatory genomic variants within the subnetwork gene set of the drug that affect therapeutic efficacy or adverse events. Methods of using these results to characterize drug responses in humans for clinical decision support, drug reuse including development of novel concomitant therapies, and for drug target discovery in silico are described.

Background

Spatial drug epigenome, super enhancer and topologically associated domain

New insights into the architecture and dynamics of non-coding regulatory genomes have changed the traditional view of pharmacodynamics and pharmacokinetics. The non-coding regulatory genome whose variation affects drug response in the human body is hereinafter referred to as the "drug epigenome". The drug epigenome can be defined as the active noncoding domain of the human genome, which consists of spatial, temporal and mechanical regulatory mechanisms of gene regulation in response to xenobiotic stimuli. It contains regulators of gene expression, including enhancers, promoters, and regulatory RNAs, and is characterized by a hierarchy of nicked transcriptional domains in which variations greatly affect drug responses in the human body. Transcriptional control consists of a canonical 3D structure comprising an enhancer-promoter pair, a super enhancer, a transcriptional center, mRNA splicing factors, a Topologically Associated Domain (TAD), and a Lamina Associated Domain (LAD). A specific restricted set of canonical 3D structures is activated or inhibited in a cell-type specific manner. Drug-disease networks are tightly coupled such that gene variants significantly associated with a disease are identical to, or exist within, the same regulatory network that determines the outcome of a drug-based therapy. Thus, mutations that disrupt the transcriptional spatial hierarchy within euchromatin not only convey disease risk, but also the attendant variability in drug response. An approach containing disease risk, drug response, and concomitant adverse event variants is a fertile network for discovering new drug targets using genotype/phenotype-guided computational strategies. These insights better inform patient treatment options based on emerging pharmacological bases of drug responses and adverse drug events. Examples of future therapeutic strategies involving combinatorial drug design targeting one or more pharmacogenomic networks, integrated multi-scale analytical approaches derived from different data type sets to enhance hierarchical-based drug discovery through molecular-by-molecular environmental modification of drug epigenomes, altering the synthetic editing of non-coding regulatory elements that convey drug therapy resistance, and developing transcription factor-like molecules to cell reprogram tissue injury and atrophy.

It is important to note that the highly significant SNP trait associations examined for hundreds of thousands of people, both from Genome Wide Association Studies (GWAS), phenome wide association studies (PheWAS), and other biobank patient data (including SNPs conveying disease risk), and individual responses to specific drugs are present in non-coding genomic regulatory elements called enhancers. In many cases, enhancers target gene promoters or regulatory RNAs within the same TAD, and may be controlled by larger regulatory elements known as super enhancers.

Human genes that play critical roles in health, disease state, and drug response are often regulated by long DNA elements spanning 2 or more TADs (referred to herein as super-enhancers) referred to as "super-enhancers" or "tensile enhancers". Super enhancers are enhancer clusters that are occupied by an abnormally high density of interacting factors and activate differential transcription (also referred to as gene expression) at a higher frequency than typical enhancers exhibit. Super enhancer super enhancers are nucleolar assemblies that represent macromolecular condensates that concentrate transcriptional regulation within the nucleus of a cell and compartmentalize it. Super enhancer super enhancers occupy known genomic positions across multiple TADs and LADs in a cell and developmental specific manner.

Super-enhancer super-enhancers are altered and disrupt their regulation of genes and RNA resulting in the elimination or alteration of chromatin loops between enhancer-promoter or promoter-promoter pairs, and/or mutations that break the boundaries of TAD or scatter the inhibitory subset of TAD, known as LAD, have profound effects on the variability of drug responses and the incidence of adverse drug events in the population.

The largest magnitude of pharmacogenomic effect was found in patients with SNPs, which are single base changes that disrupt the super-enhancer, leading to life-threatening acute adverse drug events. Examples include clozapine (clozapine) -induced agranulocytosis/granulocytopenia and Stevens Johnson syndrome or toxic epidermal necrolysis caused by carbamazepine (carbamazepine), lamotrigine (lamotrigine), phenobarbital (phenobarbital), allopurinol (allopurinol), non-steroidal anti-inflammatory agents and certain other drugs. These drug adverse responses were severe enough that countries such as singapore and taiwan require patients to test for the presence of these SNPs prior to administration of these drugs.

Super enhancers are responsible for the identification of different cell types during a given development and serve as a platform for binding neural specific transcription factors and mediator complexes in tissues such as the brain. It represents a non-traditional pharmacodynamic target, and its involvement in differential neurogenesis in the adult brain is also a mechanism by which histone deacetylase inhibitors exert their effects in the CNS. Similarly, the unconventional interpretation of drug responses and mitigation of Single Nucleotide Polymorphisms (SNPs) from GWAS and PheWAS significantly improves the understanding of the way how mutational perturbations to the molecular physiology of cells lead to human pharmacogenomic variation.

The spatial hierarchy of the transcribed tissue was first determined by a chromatin conformation capture method. The chromosomes fill most of the available volume of the nucleoplasm as a chromosome map (CT) and contain restricted a and B compartments consisting of euchromatin and heterochromatin, respectively. Typically, compartment a contains euchromatin and more active gene transcription, and compartment B corresponds to heterochromatin and is gene-poor. Compartment B incorporates the LAD located at the periphery of the nucleus. These are features that are unique to chromosome layout and appear to be largely invariant to chromatin organization, as they are not destroyed when the TAD or LAD tissues are destroyed using genome editing methods. The A and B chromatin compartments of CT contain approximately 2,450 TADs, with the linear sequences having an average length of 100Kbp (kilobase pairs) to 5Mbp (megabase pairs). TAD was first characterized using a chromatin conformation capture method such as Hi-C, allowing high resolution studies of the organization of enhancer-promoter loops and TAD boundary proteins within TAD.

Chromosomes are contained within the nucleoplasm of differentiated cells as large rope helices of genomic DNA enclosed in chromatin. CT exists in 3D space where spatial proximity and chromatin state determine regulatory interactions, rather than distance as measured in linear DNA sequences. Although there is largely no overlap of the CTs, there are a variety of spatial interactions between different CTs that can work. These comprise a complex transcription center consisting of multiple genes, regulatory elements including enhancers and promoters, and functionally related DNA binding proteins, such as transcription factors. Trans-interactions include spatial contacts between chromosomes involving enhancer-promoter interactions or, in some cases, promoter-promoter pairs.

Drug-restricted TAD alteration in a cell-type specific manner

Most human genomes are divided into approximately 2,450 basic transcription units called TADs, but about 5% of the expressed genes and functional long non-coding RNAs are not present within these bound 3D structures. TAD is delineated by a restricted boundary, usually contains multiple functionally related genes controlled by enhancers within TAD, and is invariant across all cell types studied to date. Unless the TAD boundary is disrupted by SNPs or other genetic variants, few enhancers cross the TAD boundary in most cases. The difference in gene expression between different cell types is a function of whether TAD is activated or inhibited in that cell type. TAD exhibits specific histone modifications, is a unit of DNA replication timing, and a specific and restricted TAD set and its trans-interacting TAD comprise both drug-responsive and hormone-responsive co-regulatory modules. The boundaries of TAD are relatively invariant between different human cell types. The strength of the TAD boundary can be classified into 5 different domains based on the amount of CTCF bound to the boundary and whether the super-enhancer is co-located on the TAD boundary.

Regulatory pharmacogenomics determination pharmacogenomics network

From recent studies, some basic principles of the drug epigenome have emerged: (1) results from GWAS and PheWAS indicate that more than 90% of the causal Single Nucleotide Polymorphisms (SNPs) are located within regulatory enhancers and about 5% within protein coding exons; (2) in adult humans, chromatin contact between an enhancer and a promoter or a coordinated promoter always precedes both gene transcription and alternative splicing of mRNA encoding a protein; (3) histone modifications indicate the regulatory state of any given genomic regulatory element or gene; and (4) in all cases studied to date, causal genetic variants exhibit allelic specificity, whether the cell is diploid, tetraploid or octaploid. Numerous studies have shown that it is possible to predict whether genetic mutations (such as SNPs located within enhancers) are causal using machine learning algorithms that have been trained on DNase I hypersensitivity, which is indicative of allele specificity, and other properties of epigenomes that contain histone modifications associated with enhancers and promoters. The accuracy of the clinical utility of these machine learning applications has been verified using known causal SNPs and comparing them to the output of these software programs.

Recent studies have shown, from laboratories and those other researchers, that there are a new class of pharmacodynamic and pharmacokinetic master regulator networks in chromatin that function to activate and inhibit large interconnected gene sets that are in contact in the chromatin space. The controllers of these pharmacogenomic regulatory networks represent a new class of druggable targets in the human body, different from the previous generation of epigenetic drugs, consisting of writers, readers, and erasers.

Enhancer and super enhancer SNPs associated with disease risk and drug response are key to the discovery of pharmacogenomic networks

Causal mutations such as SNPs found within enhancers, promoters and splice sites greatly alter chromatin state and can be used as "data probes" for the discovery of drug networks within the 3D spatial environment of chromatin located within the nucleus. Despite the published literature describing methods for developing drug networks even in complex tissues like the human brain, the current gene-gene and protein-protein regulatory pathways are problematic in that: (1) it is based on the hypothesis that mutations within protein-coding genes and protein-coding exons represent the most major biologically relevant mechanisms, and/or (2) it looks for similarities that mimic the structural and catalytic properties of new compounds of FDA-approved drugs for a given indication or matches tissue-specific gene expression patterns to those of FDA-approved drugs for a given indication. Recent studies have shown that none of these assumptions is very accurate for the discovery of new psychotropic drug candidates that provide better efficacy and fewer side effects than existing drugs. First, the most important SNP trait association of disease risk and drug response alters the function of enhancers located within non-coding genomes rather than proteins, and genetic variants located within introns of a gene disrupt this intragenic enhancer which may or may not regulate the expression of the gene in which it is located. SNPs located within protein coding exons often disrupt alternative splicing of mRNA or may disrupt enhancers, making any method of predicting a priori that missense SNPs will alter protein products inaccurate. Similarly, there are many functional RNAs in the human genome, including long non-coding RNAs that are not translated into protein. Second, the "guilt-by-assembly" method, as used by the integrated network-based cell feature Library (LINCS) project, is based entirely on gene expression profiling in cell lines as a surrogate to discover new drugs for human tissues, such as the brain. The complexity of this human tissue requires a more detailed, comprehensive interrogation method than that provided by surrogate pregnancy methods using cell line-dependent "shot-gun" expression profiles.

Disclosure of Invention

A method and system for detecting regulatory drug networks in the human body using bioinformatics and computational methods such as machine learning and deep learning. The basis of these approaches is the ability to uncover previously unidentified pharmacogenomic networks by interrogating pharmacogenomic regulatory interactions embedded within the functional three-dimensional (3D) topology of the human genome using mutations that stratify drug responses in large populations. These spatial regulatory interactions provide the framework for pharmacogenomic networks for most psychotropic and antineoplastic drugs.

There are now a number of existing data that can be used to computationally map drug pathways to replace additional experiments in animal and cell models, and without relying on the use of complex probabilistic reasoning methods. These knowledge-based methods described herein can be used to reconstruct pharmacogenomic networks of drugs acting on different cell types and tissues, and deconstruct these networks into components of post-hoc validation using bioinformatic analysis that mediates different on-target (on-target) and off-target (off-target) mechanisms of drugs.

These methods differ from those that require experimental perturbation of the biology of cells or tissues following drug exposure, or that rely entirely on the centrality of the learning machine for pathway mapping. An important part of the process of determining whether a single nucleotide polymorphism (SNP: which may be a single base pair change or a short insertion/deletion) is significantly associated with a particular drug response is the use of different machine learning algorithms to determine possible mechanistic causal relationships. Nevertheless, the main mapping method is based on 3D genomic structures and existing knowledge bases derived from multiple public data sources and/or from experimental or proprietary data.

Fig. 1D illustrates an exemplary model of how the system uses machine learning and deep learning to integrate and process multi-scale data for pharmacogenomic network reconstruction. The strategy for drawing the drug network provides insight on the on-target and off-target effects of mechanistic science, and lays a foundation for subsequent preclinical research;

figure 1E illustrates a method for detecting a pharmacogenomic network in a human that can be performed by a server device. The first step of the method involves extracting important SNPs associated with a specific drug response. Most of these SNPs have been published in Genome Wide Association Studies (GWAS) and phenome wide association studies (PheWAS), and there are also a large number of unbiased, peer-reviewed scientific publications from which such data can be obtained. To improve the accuracy of the location of the SNP, it was processed by the server device using the automated Pharmacogenomic Informatics Pipeline (PIP) described in fig. 4B, 4D, and 4E, referenced in us patent application No. 15/977,347 filed 2018, 5, month 11, which is incorporated herein by reference. Once the interpolation and annotation is performed to further characterize the SNP in its organizational context, a variety of accurate and validated machine learning algorithms trained in causal disease SNPs are applied to determine possible mechanical causal relationships. In addition, missense SNPs, synonymous SNPs and SNPs located within exons, which may be splice site donors or acceptors, are characterized using machine learning. The output of this pipeline is a set of "allowed" candidate SNPs that have been shown to stratify drug responses within a population to a particular drug of interest. The next method step performed by the server device comprises performing spatial genome classification using these causal SNPs to localize their target genes within the same TAD as these SNPs, with an enhancer, and using these enhancer SNPs to localize statistically significant spatial contacts within the genome that are top-ranked (e.g., top three) of their resident TADs by analyzing the data set generated using the chromosome conformation capture method (most typically generated by the Hi-C method). If a causal enhancer SNP resides in a TAD with an empirically determined strong border of intensity III-V, characterized by a TAD border containing genes involved in drug uptake, distribution, metabolism and excretion (ADME), all genes within the TAD controlled by the same enhancer are conserved for further evaluation. Similarly, if the top statistically significant contacted "trans TAD" in the spatial genome contains a gene controlled by an enhancer active in the same cell type and/or tissue in which the drug acts, it will also be saved for further evaluation.

The candidate gene set containing both endo-TAD and trans-TAD genes is then evaluated for known network connectivity using pathway analysis from, for example, third party software. Genes that do form statistically significant interconnected pathways, most commonly determined using Fisher's exact test, expressed for a drug of interest in a tissue of interest comprise a preliminary set of candidate spatial network genes. Genes that are not significantly interconnected with other genes are discarded. This includes a preliminary set of genes for a spatial network of a particular drug.

Knowledge-based semi-automated and automated remediation is then performed on this set of spatial networks comprising a particular drug to evaluate the corresponding added or removed genes. First, each member of the gene set is thoroughly examined in the context of its defined function, including the main scientific publications that evaluate its function in the context of the particular drug of interest. Next, the effect on the known efficacy and adverse event mechanisms of a particular drug of interest was evaluated for the entire set of known mutations within each gene, defined as +10 kilobases (bp) in linear distance from its transcription start site and its stop codon. Mutations comprise SNPs, variable numbers of tandem repeats, duplications, and large insertions or deletions. In this context, any functional relationship of a physiological process that is associated with the efficacy or adverse event of a particular drug of interest is included in the evaluation process. This is not limited to the effect of pharmacogenomics on specific drug responses. Third, in complex tissues such as the human brain, the expression pattern of each gene is compared to neuroanatomical substrates, where the specific drug of interest is known to act according to other studies. For example, in reconstructing a ketamine (ketamine) spatial network, data sets from 24 functional neuroimaging studies were examined to determine which brain regions were metabolically active following administration of ketamine in humans. Each gene in the preliminary gene set of the spatial network of ketamine was examined to see if its expression in the human brain overlaps with a consensus neural map derived from 24 functional neuroimaging studies, detailing the neuroanatomical substrates on which ketamine acts in the human brain. To accomplish this task, the neuroanatomical neuropograms of each gene in the human Brain were examined for microarray expression and in situ hybridization results from the human Brain map of the Allen Brain Science Institute (Allen Brain Science Institute) and RNA-seq results from the GTEx project of the National Institutes of Health (National Institutes of Health). Genes whose expression pattern does not fit the consensus neuroanatomical atlas are discarded.

For the pharmacogenomic netgenomics gene set, pathway analysis (e.g., by third party software) is used to re-evaluate whether each of the genes has a known network connection. Genes that do form statistically significant interconnected pathways, most commonly determined using fisher's exact tests, for drug expression of interest in a tissue of interest comprise a preliminary set of candidate spatial network genes. Genes that are not significantly interconnected with other genes are discarded. This includes the last gene set of the specific drug spatial network.

The next step of the method comprises applying iterative gene set optimization tools and algorithms for organizing spatial network genes into functional subsets of genes, some of which include sub-networks of drug efficacy and drug adverse events within a larger gene set. This involves a measure of similarity of input molecules derived from one or more of a number of data sources relating to the mechanism of action of a particular drug, converted to standardized human gene nomenclature, and compared to genes of a pharmacogenomic network. The output of this process is the entire set of genes of the spatial network of a particular drug, organized into its component sub-networks, including sub-networks of efficacy and adverse events.

The next step of the method involves providing scientific validation of the spatial network of specific drugs organized into their constituent sub-networks using third party software applications such as bioinformatics and biometrics. These include examples of mutational functional impairment of top-ranked (e.g., top five) statistically significant terms from Gene Ontology (Gene Ontology) or drug databases such as MedDRA, top-ranked (e.g., top five) canonical pathways determined by pathway analysis (such as by commercial or open source pathway analysis software programs), top upstream xenobiotic modulators, and spatial networks and their sub-networks annotated with statistically significant SNP trait associations from GWAS and PheWAS.

After the verification is performed, the spatial network of the particular medication and its constituent sub-networks may be stored in a database and provided to the client device for display.

Spatial networks of drugs and their constituent sub-networks can be applied in several contexts. For example, different examples are presented in pharmacogenomic decision support for drug selection, drug reuse, and drug target discovery in silico. One example of clinical decision support is a method of matching the sub-networks of efficacy and adverse events of a reference pharmacogenomics network and its database selected from such a spatial network to the sub-networks of patient-specific drug efficacy and adverse events. This comparison uses a method in deep learning in which a synergy training of efficacy metrics is conducted between the reference and patient sub-networks and the pattern matching scores. The outputs are the individual drug efficacy similarity score and the drug adverse event sub-similarity score. It should be noted that those trained in the art will recognize that reference to pharmacogenomics networks and their component efficacy and adverse event sub-networks do not represent optimal profiles. Rather, it reflects the entirety of the mechanism of action of the drug, covering both the best and worst effects that the drug may have on an individual patient.

An example of drug discovery in silicon is the selection of the gene member PPP1R1B gene in a gene set in the ketamine spatial network, and controlled by the same enhancer that controls the gene NEUROD2 (a gene whose protein product is involved in neurogenesis) and in significant spatial contact with the trans TAD containing genes DRD2 and ADORA 2A. Mapping the gene set interconnected with the PPP1R1B gene using the methods described herein and evaluating the gene ontology in terms of terms and associated canonical pathways is a way to show that it is very significantly involved in Central Nervous System (CNS) development, neuronal differentiation and neurogenesis. In addition, the PPP1R1B gene is expressed in a restricted collection of human brain regions containing the foretail, nucleus accumbens and the shell, as are the majority of the 24 genes that are significantly interconnected with the gene (involved in reward and addictive neuroanatomical substrates). Finally, PPP1R1B encodes a druggable phosphoprotein, which is defined as a "bifunctional signal transduction molecule". Dopaminergic and glutamatergic receptors stimulate modulation of their phosphorylation and act as kinase or phosphatase inhibitors. As a target for dopamine, this gene can be used as a therapeutic target for neurological and psychiatric disorders. This represents a potential druggable drug target identified using these methods.

The results of these methods comprise a spatial network of the drugs ketamine, valproic acid, lithium, lamotrigine, clozapine and warfarin. Post-hoc validation of these pharmacogenomic networks using bioinformatics methods and knowledge-based segmentation of their sub-networks of efficacy and adverse events using the methods of the present disclosure are provided. Details regarding specific efficacy and adverse event sub-networks are also provided to demonstrate the output of the pharmacogenomic network identification system.

Drawings

FIG. 1A illustrates a block diagram of a computer network and system on which an exemplary pharmacogenomic network identification system can operate, according to the presently described embodiments;

fig. 1B is a block diagram of an exemplary pharmacogenomics network server that can operate in the system of fig. 1A according to the presently described embodiments;

FIG. 1C is a block diagram of an exemplary client device that may operate in the system of FIG. 1A, according to the presently described embodiments;

fig. 2 illustrates an exemplary model of how the system integrates and processes multi-scale data using machine learning and deep learning for pharmacogenomics network reconstruction. The strategy for drawing the drug network provides a solution to the on-target and off-target effects of mechanistic science, and lays a foundation for the subsequent preclinical research;

FIG. 3 shows an example of a TAD containing the gene promoter, enhancer, super enhancer, architectural proteins contained within the TAD boundaries and subsequent chromatin looping from the promoter to different exons during alternative splicing of the gene;

FIGS. 4A and 4B show the property of drug expansion of adjacent TADs that is effected by activating the enhancer and/or super enhancer to cause differential gene expression. FIG. 4C demonstrates that TAD structure of the human genome provides more accurate information about the target gene localization of enhancers and/or super enhancers compared to traditional measures of linkage disequilibrium within the population;

figure 5A shows a "yarn ball" model of human genomic chromatin organization within the nucleus, including chromatin spatial interactions. FIG. 5B shows a simple drug network in which 3 super enhancers regulate 6 TADs and 4 TADs lack super enhancer regulation, and their trans-interactions in the spatial genome following "yarn ball" exposure to drug;

figure 6 shows a simple example of how a SNP located within the enhancer in the network may disrupt contact of the enhancer with one of its target gene promoters in the TAD, resulting in a drug adverse event in the patient within the drug response cohort. Figure 6A shows how metrics can be obtained from chromatin space interaction groups in three dimensions using different laboratory methods and the data analyzed in the form of a 2-dimensional map of enhancer-gene promoter interactions. Fig. 6B depicts how a SNP may disrupt the chromatin loop between an enhancer and one of the two gene promoters that it regulates within TAD. This disruption removes the spatial linkage between the enhancer and gene promoter 1, resulting in a dysregulation of gene 1, resulting in the patient and his cohort responding to administration of the particular drug of interest;

FIG. 7 illustrates the characteristics of a spatial genome, comprising several enhancers in each of the TADs located within non-coding genomic DNA (i.e., intergenic or introns) that selectively activate or inhibit a particular functionally related gene within the TAD;

figure 8A demonstrates the property of significant association between ADME genes and super enhancers in humans, and figure 8B shows that the association between non-coding variations within a super enhances those that can significantly alter psychotropic drug responses;

figure 9 shows an example of a comparison of SNP rs12967143-G (intragenic enhancer located in TCF4 gene) with the significance test results of other GWAS SNPs as described using numerical outputs from six different machine learning algorithms used in the analysis and among various neural and non-neural cell types (. p.ltoreq.0.05;. p.ltoreq.0.01; ANOVA);

figure 10 demonstrates that TAD containing PK and HLA gene clusters have strong TAD boundaries and are associated with important biological processes as determined by gene ontology;

figure 11 illustrates a flow chart representing a method for generating a reconstructed pharmacogenomic network comprising a human pharmacogenomic SNP input filter, a pharmacogenomic network reconstruction engine, and an iterative gene set optimization engine for a drug of interest, and a corresponding sub-network, an output drug efficacy and adverse event sub-network;

FIG. 12 illustrates a flow chart representing an exemplary method for iterative gene set optimization to deconstruct a drug pharmacogenomic network into sub-networks;

FIG. 13 sets forth a flow chart representing an exemplary method for post-hoc validation of pharmacogenomics networks and their component sub-networks using standardized bioinformatics analysis;

FIG. 14 illustrates a flow chart representing an exemplary method for error correction of a pharmacogenomics network and its component efficacy and adverse event sub-networks using individualized patient response data;

figure 15 shows a flow chart representing an exemplary method for matching drug efficacy and adverse events of a patient with drug efficacy and adverse events of a reference pharmacogenomic network using a similarity score for optimizing drug selection in clinical decision support;

fig. 16A shows a flow chart representing an exemplary method of drug target identification and drug retargeting in silicon of druggable target PPP1R 1B. Figures 16B and 16C also show graphical representations of some of the features in the characteristics of the druggable target PPP1R1B within sub-network 2 of the neuronal development and antidepressant mechanisms of the ketamine spatial network demonstrating the behavior of the ketamine pharmacogenomic network as determined from post-hoc validation of the ketamine pharmacogenomic network and its sub-network of efficacy and adverse events. Figure 16D shows gene expression data for key pharmacogenomic efficacy genes in relevant brain tissue regions;

fig. 17A shows a general topological model of CNS and peripheral drug responses, including chromatin remodeling, PK/hormone modulation, efficacy, Adverse Events (AEs), systemic PK, and systemic AE and immune system responses. Figure 17B shows four pharmacogenomic network topology models defining psychotropic and anti-neoplastic drug responses and an exemplary set of their constituent sub-networks and these sub-networks and exemplary drugs conforming to these topologies. These topologies are used by the systems described herein;

figure 18 shows a graphical depiction of a valproic acid pharmacogenomic network and its component subnetworks in the human brain, including chromatin remodeling, efficacy, adverse events, and hormone control and pharmacokinetics, using the methods and systems of the present invention;

figure 19A shows the most significant disease annotation for valproic acid pharmacogenomic networks. Figure 19B shows the top 10 drugs as upstream modulators of valproic acid pharmacogenomic networks. Figure 19C illustrates a topological model that most accurately fits a valproic acid drug genomic network;

figure 20 shows an example of a valproic acid pharmacogenomic adverse event subnetwork, and post hoc bioinformatics analysis indicates that the valproic acid pharmacogenomic adverse event subnetwork is significantly associated with cancer, severe psychological disorders, cognitive impairment, gastrointestinal disorders, lymphoproliferative disorders, motor problems including tremor, and alopecia;

figure 21 shows an example of a valproic acid pharmacogenomic neurogenesis subnetwork, and post hoc bioinformatic analysis indicates that valproic acid pharmacogenomic neurogenesis is associated with neuronal number, morphogenesis, neuronal cell proliferation, neuronal differentiation, differentiation of embryonic tissue, epilepsy or neurodevelopmental disorders, cognitive impairment, mood disorders, Alzheimer's disease or frontotemporal dementia and migraine;

figure 22 demonstrates an example of disease risk and pharmacogenomic SNPs from GWAS that can be used to determine the predisposition of an individual patient to experience an adverse event following valproic acid therapy as indicated in figure 22A or efficacy response efficacy as shown in figure 22B;

FIG. 23 illustrates the overlap of the output using this system and method with 4 other experiments and existing data sources, including genes significantly differentially expressed by porcine (wild boar) brain following peripheral administration of 150mg/kg valproic acid, and including the Ingenity Path Analysis^TMDrug databases of KEGG, drug central, drug bank and LINCS. Note that this system outputs the greatest number of common valproic acid-induced genes compared to any of the other 2 comparisons;

figure 24 lists the genes contained within the chromatin remodeling subnetwork of the valproic acid pharmacogenomic network;

figure 25 lists the genes contained within the neuroplasticity and efficacy sub-networks of the valproic acid pharmacogenomic network;

figure 26 lists the genes contained within the adverse event sub-network of the valproic acid pharmacogenomic network;

figure 27 lists the genes contained within the pharmacokinetic and hormonal subnetworks of the valproic acid pharmacogenomic network;

FIGS. 28A-28I show selected chromatin space contacts of valproic acid pharmacogenomic networks and their functional networks as determined by chromosome conformation capture in human neurons using the Hi-C method;

figure 29A presents the most significant disease annotation for ketamine pharmacogenomic networks. Figure 29B demonstrates the first 5 drugs as upstream modulators of the ketamine pharmacogenomic network. Figure 29C shows the topological model most accurately fit to ketamine pharmacogenomic networks;

figure 30 demonstrates an example gene set enrichment of the output of a gene set optimization engine that distinguishes 2 significantly different sub-networks within 3 sub-networks including ketamine pharmacogenomics networks in the human brain. Figure 30A is a ketamine pharmacogenomic glutamate receptor subnetwork responsible for adverse events associated with drug as well as neurotransmission. Figure 30B is a ketamine pharmacogenomic neuroplasticity subnetwork mediating the antidepressant response of ketamine;

figure 31 shows an example of ketamine pharmacogenomic glutamate receptor subnetwork, and post hoc bioinformatic analysis indicates that the ketamine pharmacogenomic glutamate receptor subnetwork is significantly associated with Adverse Events (AEs) as follows: cognitive impairment, bipolar disorder, postoperative delirium, schizoaffective disorder, schizophrenia, noncancerous pain, postoperative pain, emesis, heartburn and unconsciousness;

figure 32 presents an example of ketamine pharmacogenomic neuroplasticity sub-networks, and post hoc bioinformatic analysis indicates that ketamine pharmacogenomic neuroplasticity sub-networks are significantly associated with emotional behavior, morphological abnormalities of the nervous system, morphological abnormalities of the brain, depression, anxiety, and morphological abnormalities of neurons;

figure 33 demonstrates an example of a disease risk SNP from GWAS as shown in figure 33A that can be used to determine an individual patient's propensity to experience an adverse event following ketamine therapy or the efficacy of the antidepressant response shown in figure 33B;

figure 34 lists the genes contained within the neuroplasticity and efficacy sub-networks of the ketamine pharmacogenomic network;

figure 35 lists the genes contained within the chromatin remodeling and adverse event subnetworks of the ketamine pharmacogenomic network;

figure 36 lists the genes contained within the pharmacokinetic and hormone sub-networks of the ketamine pharmacogenomic network;

figures 37A-37G demonstrate selected chromatin space contacts of the entire ketamine pharmacogenomic network as determined by chromosome conformation capture in human neurons using Hi-C;

figure 38 demonstrates a significant overlap between neuroanatomical distributions of gene expression data within the ketamine pharmacogenomic network and localization results obtained from 24 neuroimaging studies from consensus brain maps showing which brain regions were first affected by ketamine;

figure 39 demonstrates an example of the beneficial combinatorial mechanisms and therapeutics found using methods of this system in a combinatorial fashion using valproic acid and ketamine in H3K9 acetylation and deacetylation, respectively, resulting in neurogenesis and neural differentiation;

figure 40 demonstrates a complementary pharmacogenomic network of valproic acid (figure 40A) and a pharmacogenomic network of ketamine (figure 40B) showing neurogenesis and neural differentiation, respectively;

figure 41 shows the combined effect of valproic acid and ketamine pharmacogenomic networks in neurogenesis, neuronal proliferation and terminal neuronal differentiation;

figure 42A shows the most significant disease annotation for the lithium pharmacogenomic network. Figure 42B shows the top 5 drugs as upstream modulators of the lithium pharmacogenomics network. Figure 42C shows a topology model that fits most accurately to a lithium pharmacogenomics network;

figure 43 shows high resolution compartmentalization of a gene set sub-network as one example of an output using this system for lithium pharmacogenomics networks;

figure 44 lists the genes contained within the chromatin remodeling subnetwork of the lithium pharmacogenomics network;

figure 45 lists the genes contained within the neuroplasticity subnetwork of the lithium pharmacogenomics network;

figure 46 lists the genes contained within the efficacy sub-network of the lithium pharmacogenomic network;

figure 47 lists the genes contained within the drug-induced weight gain (adverse event) subnetwork of the lithium pharmacogenomics network;

figure 48 lists the genes contained within the drug-induced tremor (adverse event) subnetwork of the lithium pharmacogenomics network;

figure 49A shows the most significant disease annotation for lamotrigine pharmacogenomic networks. Figure 49B shows the first 5 drugs as upstream modulators of lamotrigine pharmacogenomic networks. Figure 49C shows a topological model that most accurately fits lamotrigine pharmacogenomic networks;

figure 50 shows an example of lamotrigine pharmacogenomic adverse event subnetworks as the output of this system;

figure 51 shows an example of lamotrigine pharmacogenomic neuroplasticity and efficacy sub-networks as the output of this system;

figure 52 lists the genes contained within the chromatin remodeling subnetwork of the lamotrigine pharmacogenomic network;

figure 53 lists the genes contained within the neuroplasticity sub-network of lamotrigine pharmacogenomic networks;

figure 54 lists the genes contained within the adverse event sub-network of the lamotrigine pharmacogenomic network;

figure 55 lists the genes contained within the pharmacokinetic subnetwork of the lamotrigine pharmacogenomic network;

figure 56A shows the most significant disease annotation for clozapine pharmacogenomic networks. Figure 56B shows the first 5 drugs as upstream modulators of clozapine pharmacogenomic networks. Figure 56C shows the topological model most accurately fit to the clozapine pharmacogenomic network;

figure 57 shows an example of a sub-network of clozapine pharmacogenomic adverse events as the output of this system;

figure 58 shows an example of clozapine pharmacogenomic neuroplasticity and efficacy sub-networks as the output of this system;

FIG. 59 lists the genes contained within the chromatin remodeling subnetwork of the clozapine pharmacogenomic network;

FIG. 60 lists the genes contained within the neuroplasticity subnetwork of the clozapine pharmacogenomic network;

figure 61 lists the genes contained within the adverse event sub-network of the clozapine pharmacogenomic network;

figure 62 lists the genes contained within the pharmacokinetic subnetwork of the clozapine pharmacogenomic network; and is

Figure 63 shows a warfarin pharmacogenomics network that is not drawn as any of the network topologies for psychopharmaceuticals as shown in figure 11. The warfarin pharmacogenomic network is shown in fig. 63A, and the corresponding gene set enrichment characteristics are shown in fig. 63B. Fig. 63C shows gene set enrichment of the warfarin anti-coagulation subnetwork, and fig. 63D shows gene enrichment of the warfarin bleeding and vascular occlusion subnetwork.

Detailed Description

Although the following text sets forth a detailed description of numerous different embodiments, it should be understood that the legal scope of the description is defined by the words of the claims set forth at the end of this disclosure. The detailed description is to be construed as exemplary only and does not describe every possible embodiment since describing every possible embodiment would be impractical, if not impossible. Numerous alternative embodiments could be implemented, using either current technology or technology developed after the filing date of this patent, which would still fall within the scope of the claims.

It will also be understood that, unless the sentence "as used herein is used in this patent, the term '______' is defined herein to mean … …" or a similar sentence with a definite definition of the term, there is no intent to limit the meaning of that term, whether by express or by implication, beyond its plain or ordinary meaning, and that this term should not be interpreted as limited in scope based on any statement made in any section of this patent (other than the language of the claims). To the extent that any term recited in the claims at the end of this patent is referred to in this patent in a manner consistent with a single meaning, that is done for sake of clarity only so as to not confuse the reader, and it is not intended that such claim term by limited, by implication or otherwise, to that single meaning. Finally, unless a claim element is defined by reference to the word "means" and a recited function without any structure, it is not intended that the scope of any claim element be construed in accordance with the application of 35u.s.c. § 112 (sixth paragraph).

This section presents a detailed description of a pharmacogenomic network identification system and its application to drug phenotypic decision support found in silicon for drug selection and pharmacodynamic drug targets within biological pathways. A description of methodology and its application in clinical medicine and drug research is presented first, followed by several exemplary illustrations of pharmacogenomic networks. The examples are non-limiting and related variations of the method will be apparent to those skilled in the art and are intended to be encompassed by the appended claims.

The pharmacogenomic network identification system generates a model of the pharmacogenomic regulatory network and its constituent subnetworks using a contemporary knowledge base comprising the functional topology of the pharmacogenomic genomic architecture, 3D molecular circuits within chromatin that control gene expression and mRNA splicing, and drug-specific geometric expansion and contraction of TAD and its super-enhancer regulated pharmacogenomic linkages that affect enhancer-promoter and promoter-promoter interactions.

As shown in figure 8, the nature of these interactions comprises a significant association of genes encoding proteins involved in the absorption, distribution, metabolism and excretion (ADME) of xenobiotic drugs-an example consists of known mutations in the super enhancer GH06J032184 responsible for a drug adverse event known as neutropenia occurring in certain individuals following treatment with the antipsychotic drug clozapine.

The reconstituted pharmacogenomic networks described herein are indistinguishable from those mediating disease etiology, thus providing another avenue for studying pharmacological mechanisms of action. Based on the responsive chromatin plasticity in which it is embedded, the pharmacogenomic network adapts over time to intrinsic and extrinsic stimuli, which explains pharmacogenomic variation in humans. This determines the response of individual patients to drugs, including adverse drug events, and examples of such variations caused by different ratios of sub-networks within a pharmacogenomic network in a patient will be provided as output examples and methods of such systems.

In general, the techniques for identifying a pharmacogenomic network of drugs can be implemented in one or several client devices, one or several network servers, or a system comprising a combination of these devices. However, for clarity, the examples below focus primarily on embodiments where the pharmacogenomics web server obtains SNPs from human clinical studies that have been demonstrated to be significantly associated with responses and adverse events with respect to the particular drug of interest, or which may contain disease or trait risk SNPs. The pharmacogenomics web server compares the SNP to SNPs reported from Genome Wide Association Studies (GWAS), bio-pools, phenome wide association studies (PheWAS), and other candidate gene studies to identify additional SNPs linked to the SNPs using the characteristics of the three-dimensional (3D) genome topology used to generate the set of permissive candidate variants.

The pharmacogenomics network server then performs bioinformatics analysis on each of the licensed candidate variants to filter the set of SNPs into a subset of intermediate candidate variants based on regulatory function, variant dependency, presence of target gene relationship of the licensed candidate variants, and/or whether the licensed candidate variants are non-synonymous or synonymous coding variants that do not affect the protein but are involved in regulation of gene expression. In addition, the pharmacogenomics network server performs pathway analysis on the target genes associated with the subset of intermediate candidate variants to filter the target genes to identify a set of genes that are causally related to the particular drug.

The pharmacogenomic network server then identifies a pharmacogenomic network for the particular drug of interest based on the identified gene set, and provides an indication of the pharmacogenomic network for display to the client device. For example, the indication of the pharmacogenomic network may comprise the name of the drug of interest, the name and/or graphical depiction of each of the genes in the pharmacogenomic network, and the name and/or graphical depiction of each of the sub-networks and the genes in each of the sub-networks within the pharmacogenomic network.

The pharmacogenomics web server can analyze the data described herein, such as genomic data and spatial contact data, using various machine learning techniques including, but not limited to, regression algorithms (e.g., ordinary least squares regression, linear regression, logistic regression, stepwise regression, multivariate adaptive regression splines, locally estimated scatter plot smoothing, etc.), example-based algorithms (e.g., k-nearest neighbors, learning vector quantization, self-organizing maps, locally weighted learning, etc.), regularization algorithms (e.g., ridge regression, minimum absolute contraction and selection operators, elastic nets, minimum angle regression, etc.), decision tree algorithms (e.g., classification and regression trees, iterative dichotomy 3, C4.5, C5, chi-square automated interaction detection, decision stumps, M5, conditional decision trees, etc.), clustering algorithms (e.g., k-means, k-medians, expectation maximization, hierarchical clustering, spectral clustering, mean shift, density-based spatial clustering for noisy applications, rank points for identifying clustering structures, etc.), association rule learning algorithms (e.g., a priori algorithm, Eclat algorithm, etc.), bayesian algorithms (e.g., naive bayes, gaussian bayes, polynomial naive bayes, mean single dependence estimators, bayesian belief networks, bayesian networks, etc.), artificial neural networks (e.g., perceptrons, Hopfield networks, radial basis function networks, etc.), deep learning algorithms (e.g., multi-layer perceptrons, deep boltzmann machines, deep confidence networks, convolutional neural networks, stacked autoencoders, generating countermeasure networks, etc.), dimension reduction algorithms (e.g., principal component analysis, principal component regression, partial least squares regression, etc.), sammon mapping (Sammon mapping), multidimensional scaling, projection tracking, linear discriminant analysis, hybrid discriminant analysis, quadratic discriminant analysis, flexible discriminant analysis, factorial analysis, independent component analysis, non-negative matrix factorization, t-distributed random neighborhood embedding, etc.), integration algorithms (e.g., enhancement, bootstrap aggregation, AdaBoost, stacked generalization, gradient enhancement machine, gradient-enhanced regression tree, random decision forest, etc.), reinforcement learning (e.g., moveout learning, Q learning, learning automata, state-action-reward-state-action, etc.), support vector machines, hybrid models, evolutionary algorithms, probabilistic graphical models, etc.

Referring to fig. 1A, an exemplary pharmacogenomic network identification system 100 identifies a pharmacogenomic network for various drugs. The pharmacogenomic network identification system 100 comprises a pharmacogenomic network server 102 and a plurality of client devices 106, 116 that can be communicatively connected via a network 130, as described below. In one embodiment, the drug pharmacogenomics network server 102 and the client device 106 and 116 may communicate via wireless signals 120 over a communication network 130, which may be any suitable local or wide area network, including a WiFi network, a bluetooth network, a cellular network (e.g., 3G, 4G, Long Term Evolution (LTE), 5G), the internet, and the like. In some cases, the client devices 106 and 116 may communicate with the communication network 130 through intervening wireless or wired devices 118, which may be wireless routers, wireless repeaters, base station transceivers of mobile phone providers, and the like. For example, the client device 106-116 may include a tablet computer 106, a smart watch 107, a network-enabled cellular telephone 108, a wearable computing device (e.g., Google Glass)^TMOr109) Personal Digital Assistant (PDA)110, mobile device smart phone 112 (also referred to herein as a "mobile device"), laptop computer 114, desktop computer 116, wearable biosensor, portable media player (not shown), tablet phone, any device configured for wired or wireless RF (radio frequency) communication, and the like. In addition, any other suitable client device that records omic data of a patient, receives a pharmacogenomic data set, or displays an indication of a pharmacogenomic network and/or subnetwork may also be served with the pharmacogenomic networkThe server 102 communicates.

Each of the client devices 106-116 can interact with the pharmacogenomic network server 102 to identify a drug of interest for determining a corresponding pharmacogenomic network. Each client device 106-116 may also interact with the pharmacogenomic network server 102 to receive indications of several pharmacogenomic subnetworks within the pharmacogenomic network and/or the pharmacogenomic network for the drug of interest. The client device 106 may present an indication for display to a healthcare professional or researcher through the user interface, such as a display showing the degree of overlap based on the drug-specific topology map shown in fig. 17 or with the similarity scores as shown in fig. 15.

In an example embodiment, the pharmacogenomics network server 102 may be a cloud-based server, an application server, a web server, etc., and includes a memory 150, one or more processors (CPUs) 142 (such as a microprocessor coupled to the memory 150), a network interface unit 144, and an I/O module 148, which may be, for example, a keyboard or a touch screen.

The pharmacogenomics network server 102 is also communicatively connected to a database 154 of genomic data including data from human clinical studies, biobanks, GWAS and PheWAS studies.

The memory 150 may be tangible, non-transitory memory, and may include any type of suitable memory module, including Random Access Memory (RAM), Read Only Memory (ROM), flash memory, other types of persistent memory, and the like. The memory 150 may store instructions for an Operating System (OS)152, which may be any type of suitable operating system, such as a modern smartphone operating system, for example, that is capable of executing on the processor 142. The memory 150 may also store instructions to the network reconstruction engine 146A, pharmacogenomic network bandwidth adjuster 146B, and gene set optimization engine 146C, for example, that are executable on the processor 142. The pharmacogenomic network server 102 will be described in more detail below with reference to fig. 1B. In some embodiments, the network reconstruction engine 146A, the pharmacogenomic network bandwidth adjuster 146B, and the gene set optimization engine 146C can be part of one or more of the client device 106, 116, the pharmacogenomic network server 102, or a combination of the pharmacogenomic network server 102 and the client device 106.

In any case, the network reconstruction engine 146A can receive a request from the client device 106, 116, or from a database of pre-existing reconstructed pharmacogenomic networks for identifying a pharmacogenomic network for a particular drug of interest. The client device 106-116 may also provide SNPs from human clinical studies, GWAS studies, PheWAS studies, etc. that have been shown to be significantly associated with responses and adverse events with respect to a particular drug of interest or disease risk SNPs from the pharmacogenomics subnetwork of the GWAS that distinguishes the drug. In other embodiments, the network reconstruction engine 146A may obtain SNPs from the database 154. Network reconstruction engine 146A then generates a method of permissive candidate variant sets based on the obtained SNPs and additional SNPs that are linked to the obtained SNPs according to TAD boundaries, either within the adjacent TAD regulated by super-enhancers or by remote trans-interactions determined by chromosome conformation capture methods. Further, the network reconstruction engine 146A performs bioinformatic analysis on each of the licensed candidate variants to filter the set of SNPs into subsets of intermediate candidate variants and performs pathway analysis on target genes associated with the filtered set to identify a set of genes causally related to a particular drug. The network reconstruction engine 146A identifies a pharmacogenomic network for the particular drug of interest based on the identified gene set and provides an indication of the pharmacogenomic network for display through the user interface of the client device 106 and 116.

The drug pharmacogenomic network server 102 may communicate with the client devices 106 and 116 over the network 130. The digital network 130 may be a private network, a secure public internet, a virtual private network, and/or some other type of network, such as a dedicated access line, a common conventional telephone line, a satellite link, combinations of these, and so forth. Where the digital network 130 includes the internet, data communications may be over the digital network 130 via an internet communication protocol.

Turning now to fig. 1B, the pharmacogenomic network server 102 can comprise a controller 224. The controller 224 may include a program memory 226, a microcontroller or Microprocessor (MP)228, a Random Access Memory (RAM)230, and/or input/output (I/O) circuitry 234, all of which may be interconnected by an address/data bus 232. In some embodiments, the controller 224 may also contain a database 239 or otherwise be communicatively connected to the database or other data storage mechanism (e.g., one or more hard disk drives, optical storage drives, solid state storage devices, etc.). Database 239 may contain data such as genomic data, pharmacogenomic network display templates, web page templates and/or web pages, and other data necessary for interaction with a user via network 130. Database 239 may contain data similar to database 154 described above with reference to fig. 1A.

It should be understood that although FIG. 1B depicts only one microprocessor 228, the controller 224 may contain multiple microprocessors 228. Similarly, the memory of controller 224 may include multiple RAMs 230 and/or multiple program memories 226. Although FIG. 1B depicts I/O circuitry 234 as a single block, I/O circuitry 234 may comprise many different types of I/O circuitry. The controller 224 may implement the RAM 230 and/or the program memory 226 as, for example, semiconductor memory, magnetically readable memory, and/or optically readable memory.

As shown in fig. 1B, program memory 226 and/or RAM 230 may store various applications for execution by microprocessor 228. For example, the user interface application 236 may provide a user interface to the pharmacogenomics network server 102 that may, for example, allow a system administrator to configure, troubleshoot, or test various aspects of the operation of the server. The server application 238 may be operable to receive a request to identify a pharmacogenomic network for a particular drug of interest, identify the pharmacogenomic network and the pharmacogenomic subnetwork for the particular drug, and transmit an indication of the pharmacogenomic network to the client device 106 and 116. The server application 238 may be a single module 238 or multiple modules 238A, 238B, such as the network reconstruction engine 146A, pharmacogenomics network bandwidth adjuster 146B, and gene set optimization engine 146C.

Although the server application 238 is depicted in fig. 1B as including two modules 238A and 238B, the server application 238 may include any number of modules that accomplish tasks related to the implementation of the pharmacogenomic network server 102. Further, it should be understood that although only one pharmacogenomic network server 102 is depicted in fig. 1B, multiple pharmacogenomic network servers 102 may be provided for distributing server load, serving different web pages, etc. These multiple pharmacogenomic web servers 102 may comprise web servers, entity-specific servers (e.g., a server for a particular drugServers, etc.), servers disposed in a retail or private network, etc.

Referring now to FIG. 1C, laptop computer 114 (or any of client devices 106 and 116) may include a display 240, a communication unit 258, a user input device (not shown), and a controller 242, such as pharmacogenomic network server 102. Similar to the controller 224, the controller 242 may contain a program memory 246, a microcontroller or Microprocessor (MP)248, a Random Access Memory (RAM)250, and/or an input/output (I/O) circuit 254, all of which may be interconnected by an address/data bus 252. The program memory 246 may contain an operating system 260, a data storage device 262, a plurality of software applications 264, and/or a plurality of software routines 268. For example, operating system 260 may comprise Microsoft WindowsOS And the like. Data storage device 262 may contain data such as application data for a plurality of applications 264, routine data for a plurality of routines 268, and/or other data necessary to interact with drug pharmacogenomic network server 102 over digital network 130. In some embodiments, the controller 242 may also include, or otherwise be communicatively connected to, other data storage mechanisms residing within the laptop computer 114 (e.g., one or more hard disk drives, optical storage drives, solid state storage devices, etc.).

The communication unit 258 may communicate with the pharmacogenomic network server 102 over any suitable wireless communication protocol network, such as a wireless telephone network (e.g., GSM, CDMA, LTE, etc.), Wi-Fi network (802.11 standard), WiMAX network, bluetooth network, etc. The user input device (not shown) may comprise a "soft" keyboard displayed on the display 240 of the laptop computer 114, an external hardware keyboard (e.g., a bluetooth keyboard) that communicates through a wired or wireless connection, an external mouse, a microphone for receiving voice input, or any other suitable user input device. As discussed with reference to the controller 224, it should be understood that although fig. 1C depicts only one microprocessor 248, the controller 242 may contain multiple microprocessors 248. Similarly, the memory of the controller 242 may include multiple RAMs 250 and/or multiple program memories 246. Although FIG. 1C depicts the I/O circuit 254 as a single block, the I/O circuit 254 may comprise many different types of I/O circuits. The controller 242 may implement the RAM 250 and/or the program memory 246 as, for example, semiconductor memory, magnetically readable memory, and/or optically readable memory.

The one or more processors 248 may be adapted and configured to execute, among other software applications, any one or more of a plurality of software applications 264 and/or any one or more of a plurality of software routines 268 residing in the program storage 246. One of the plurality of applications 264 may be a client application 266 that may be implemented as a series of machine readable instructions for performing various tasks associated with receiving information at the laptop computer 114, displaying information on the laptop computer, and/or transmitting information from the laptop computer.

One of the plurality of applications 264 may be a native application and/or a web browser 270 (e.g., Apple's) that may be implemented as a series of machine-readable instructionsGoogle Chrome^TM、Microsoft InternetAnd Mozilla) The series of machine-readable instructions is for receiving, interpreting, and/or displaying web page information from pharmacogenomics network server 102, while also receiving input from a user, such as a healthcare professional or researcher. Another application of the plurality of applications may comprise an embedded web browser 276 that may be implemented as a series of machine readable instructions for receiving, interpreting, and/or displaying web page information from the pharmacogenomics web server 102.

One of the plurality of routines may include a pharmacogenomic network display routine 272 that obtains an indication of the pharmacogenomic network and presents the indication on the display 240, the indication including a name of the drug of interest, a name and/or graphical description of each of the genes in the pharmacogenomic network, and a name and/or graphical description of each of the sub-networks within the pharmacogenomic network and the genes within each sub-network. Another routine of the plurality of routines may include a pharmacogenomic network request routine 274 that obtains a request for a pharmacogenomic network to identify a particular drug of interest and transmits the request to the pharmacogenomic network server 102.

Preferably, the user may launch the client application 266 from a client device (such as one of the client devices 106 and 116) to communicate with the pharmacogenomic network server 102 to implement the pharmacogenomic network authentication system 100. Additionally, the user may also launch or instantiate any other suitable user interface application (e.g., a native application or web browser 270, or any other of the plurality of software applications 264) to access the pharmacogenomic network server 102 to implement the pharmacogenomic network authentication system 100.

To identify the pharmacogenomic network for a particular drug of interest, the pharmacogenomic network server 102 performs a general method 180 as shown in fig. 11. In some embodiments, the method 180 may be implemented in a set of instructions stored on a non-transitory computer readable memory and executable on one or more processors on the pharmacogenomic network server 102. For example, the method 180 may be performed by the network reconstruction engine 146A, the pharmacogenomics network bandwidth adjuster 146B, and the gene set optimization engine 146C.

Fig. 2 illustrates the integration of multi-scale data and processing by the system 100. The methods and systems are based on the properties of pharmacogenomic networks consisting of enhancers and super-enhancer sub-networks that are activated or inhibited in the same human cell type in which the drug first functions. This can be done by matching the changes in gene expression caused by the drug or other therapeutic agent to a placebo or control with a higher order structure where the drug first acts, to determine if SNPs associated with the pharmacogenomic network and sub-network of the drug act in these human tissues, and to assess if enhancer and super-enhancer regulatory elements are located on restricted anatomical substrates of interest. Other methods of machine learning, deep learning, reinforcement learning, and artificial intelligence may be used to perform machine-executable steps in the system, where applicable.

A detailed overview of one embodiment of the system 100 is shown in fig. 11. This system combines an automated executable file with semi-manual treatment of a human pharmacogenomic SNP filter that includes two SNPs that can accept an impact on drug response or disease risk. Followed by a pharmaco-specific pharmacogenomic network reconstruction engine with a pharmaco-spatial network bandwidth adjuster, and then iterative gene set optimization for deconstruction of the pharmaco-specific pharmacogenomic network into component functional sub-networks.

Selection of genetic variants using human pharmacogenomic SNP input filters

At block 181, the pharmacogenomics web server 102 obtains SNPs from human clinical studies that have proven to be significantly associated with responses to the drug of interest and adverse events, or disease risk SNPs that can be used to differentiate the relative weights of expression, efficacy, and adverse event sub-networks within the patient's regulatory genome. Since the location of SNPs associated for traits under investigation has in most cases been inaccurately assigned to the nearest gene in the GWAS per linear sequence assembled for published literature and reference human genomes or nearby candidate genes, accurate localization using interpolation and annotation techniques was used to determine the actual location of reported SNPs.

The new research has several important significances for the pharmacogenomics network identification system. First, new drug target mechanisms can be identified by collecting pharmacogenomic network outputs in a training set using computer vision-based changes in 3D genomic architecture resulting from validation using deep-learned (machine-learned) permissive SNPs and using known drug-induced genome-wide changes in the genomic architecture. Second, clustering new drug target mechanisms in previously defined but not completely understood biological pathways will increase the probability of success. Third, the insight gained using the 3D genomic architecture for determining drug targets from drug response and disease risk SNPs will lead to next generation drug candidates and will greatly improve the accuracy of pharmacogenomic diagnostics.

Fig. 3 shows the properties of TAD-containing tissues comprising key enhancer and promoter transcription factors CREBBP (CEB-binding protein), EP300 (E1A-binding protein P300), POLR2A (RNA polymerase II subunit a) and YY1 (negative and positive transcription factor 1), as well as TAD border protein and chromatin loop-binding protein fibronectin and CTCF (CCCTC-binding protein). In addition to enhancer-promoter and promoter-promoter pairs, adjacent TADs contain super enhancers. Chromatin loop protein CTCF is also involved in pre-mRNA splicing as shown in FIG. 3.

Figure 4 demonstrates the nature of the effect of drugs on TAD modulated by super-enhancers, and the TAD-based localization of causal SNP targets provides higher accuracy compared to measures of linkage disequilibrium in the population. FIGS. 4A and 4B show 2 adjacent TADs regulated by the super enhancer. In fig. 4A without drug present, the super enhancer is silent and differential gene expression within adjacent TADs is minimal. Fig. 4B demonstrates activation of the super enhancer when 2 adjacent TADs are controlled in the presence of a drug, resulting in geometric expansion of TADs and concomitant increased expression of genes located within these expanded TADs. Figure 4C demonstrates that TAD organization in the spatially regulated genome provides a more accurate approach for locating gene promoter targets from GWAS and other causal enhancer SNPs studied, than those provided by traditional measures of linkage disequilibrium.

Figure 5 shows that the human genome is organised in a 3D manner similar to "yarn balls". This three-dimensional tissue changes in a dynamic manner over time, but regulatory interactions can be understood by examining the localization of TAD and its regulators (super enhancers) following drug-induced changes.

At block 182, the pharmacogenomic network server 102 evaluates the candidate causal relationship SNPs using the pharmacogenomic informatics pipeline. The pharmacogenomics informatics pipeline uses SNPs that have been reported from GWAS, biobank, PheWAS and other candidate gene studies to find genetically related permissive candidate SNPs using TAD boundaries rather than the measures of linkage disequilibrium shown in figure 4C. Enhancer-regulated SNP workflow is evaluated for DNA methylation, transcription factor binding, histone markers, DNase I hypersensitivity, chromatin state, Quantitative Trait Locus (QTL), chromatin loop-based contacts determined using chromosome conformation capture techniques such as Hi-C, and transcription factor binding site disruption using tissue-specific omics datasets that permit candidate SNPs in disease-related tissues. As shown in fig. 11, the pharmacogenomics network server 102 then evaluates the final output SNP using an open source machine learning algorithm to determine if the SNP is causal or not (block 183) and causal variants are retained for further analysis in the workflow (block 184). Exon SNPs were also evaluated as splice donors or splice acceptors using the Altrans algorithm. If it is found to be involved in alternative splicing, it is stored as such.

Fig. 9 shows an example of SNP selection for predicted causal relationships using 6 different machine learning and deep learning algorithms based on tissue-specific distributions. This shows the candidate SNP rs12967143-G, located within the intragenic enhancer in the transcription factor 4(TCF4) gene, and other GWAS SNPs described using the numerical output of the machine learning algorithm used in the analysis. P is less than or equal to 0.05; p is less than or equal to 0.01. Numerical scores from each algorithm were generated for each GWAS SNP, and were only SNPs retained for further analysis if each output scored a SNP that was predicted to be causal in SK-N-SH cells and H1 cells, but not HepG2 cells and PBMCs. The scores for each predicted causal SNP were tested independently to determine if the scores were significantly different for all human traits at p ≦ 5E-08 listed in the EBI-NHGRI GWAS catalog using ANOVA than the scores generated using 10 randomly selected GWAS SNPs. Only when a SNP meets this significance criterion, whether it is selected by the system for further analysis.

Using causal enhancer SNPs to interrogate pharmacogenomic networks

At block 185, cis-interactions within the target genes as the same TAD or adjacent TADs controlled by the same super enhancer are determined using enhancer SNPs as probes, and pharmacogenomic trans-interactions with other TADs are determined using Hi-C chromosome conformation capture and charateristic ia-PET data sets for mapping of 3D pharmacogenomic junctions (block 187) for performing cell types and tissues affected by the drug of interest. For a pharmacogenomics network, if the TAD participating in cis-and trans-interactions has a strong border as predicted by the amount of bound CTCF and/or significant association with the super-enhancer (block 186), genes are selected that comprise other functional elements as targets for long non-coding RNAs that are targets within the same TAD or adjacent TADs controlled by the same super-enhancer, or in trans-interactions, enhancers that significantly alter drug response in the human population, herein. For trans-interaction, if the TAD comprises the first 3 statistically significant pharmacogenomics contacts of adjacent TADs including the first set of pharmacogenomics TADs within the same cell and/or tissue type under the control of the same super-enhancer including the drug of interest, and selects genes within these "trans TADs" when they are under the control of the same cell and/or tissue specific enhancer under the drug of interest (block 188).

FIG. 7 shows the distribution of TAD properties in one cell type in the human genome. TADs of 98% contain known or predicted enhancers, and TADs of 40% have known super enhancers spanning adjacent TADs in the genome.

At block 189, the pharmacogenomics network server 102 evaluates the interconnectivity of the combined gene set, wherein the combined gene set is selected from the TAD first set that houses the pharmacogenomic SNPs and from the genes selected from "trans TAD," including genes controlled in concert with the cis-interacting gene first set. For example, the pharmacogenomics web server 102 can utilize, for example, Ingeneity Pathway Analysis^TMAnd third party software to check the connectivity of the combined gene set. Using fisher's right-hand side exact test, genes are placed into the preliminary gene set of the pharmacogenomic network that includes the drug of interest if the pharmacogenomic network server 102 determines that significant interconnectivity exists within the combined gene set based on published literature. Any genes that do not form a linked network are discarded as non-candidate genes for the pharmacogenomic network (block 190).

Knowledge-based revision using pharmacogenomic network bandwidth regulator

Then at block 191, manual, semi-automated or automated remediation, or a combination thereof, is performed on each gene in the genome comprising the preliminary pharmaco-pharmacogenomic network to remove genes whose function is not related to the drug of interest in the functioning cell and/or tissue type, or to add other genes that are not part of this preliminary set of pharmaco-pharmacogenomic networks in the set if judged to be affected by the drug of interest in the functioning cell and/or tissue type. The interrogation step involves defining the function of individual genes, the phenotypic consequences of mutational impairment of the genes, and the human cells and tissues expressed by the genes to determine whether they can be candidates for members of the pharmacogenomic network for the particular drug of interest.

In one embodiment, these determinations can be made using manual and semi-automated strategies, in combination with manual administration of each gene, its mutation profile, and the localization of its expression within human tissue. These are achieved through a variety of web-based search tools, including gene definitions, genome browser annotations, GWAS directories and other bioinformatic sources. For example, the pharmaceutical drug genomics network server 102 may call an Application Programming Interface (API) with executable files written in R, Python, PERL, or other programming languages to facilitate data access, data cleansing, and data analysis. This example is an enhanced model of manual remediation, but can become time-limited if many genes are present within the gene set of the pharmacogenomics network or the gene set of the sub-network, and especially if the functional genomic elements can comprise regulatory RNAs or functional RNAs such as long non-coding RNAs or if the function of the genes is poorly understood. The listing and analysis of the mutation status for a given gene (+10Kb upstream and downstream) is the easiest of the 3 interrogation steps to be performed, as these databases are the most comprehensive. There are other sources for the analysis of the tissue distribution of the expression pattern of genes. In comparing these patterns to sites at which a particular drug of interest acts, the pharmacogenomic network identification system 100 can utilize results from imaging modalities, including from radiology studies, optical microscopy analysis in pathology, and even more complex methods. In some embodiments, the pharmacogenomic network server 102 performs this analysis using machine learning techniques such as neural networks.

In another embodiment, the pharmacogenomics network server 102 can use a bayesian probability classifier based on machine learning or using bayesian probability calculations. Automated methods can be used to reduce the complexity of data analyzed from different data sources, where a functional knowledge profile of a gene, its mutation profile, and its tissue expression mapping are the inputs to a learning machine that has trained multiple such instances and tested independently to determine accuracy by enriching another set of instances. The predictive features selected by the trained neural network can be implemented on a support vector machine classifier to construct a functional and mutation prediction model of the gene, with subsequent machine states determining the adequacy of statistical fit to the pharmacogenomics network.

In some scenarios, machine learning tends to over-fit, outputting false positives or false negatives. In another embodiment, the pharmaceutical pharmacogenomics network server 102 can perform semi-automated and naive bayes classification using machine learning in parallel to sharpen the accuracy of the final output.

The pharmacogenomic network server 102 and, more particularly, the pharmacogenomic network bandwidth adjuster 146B can perform knowledge-based governance using the following steps. First, the pharmacogenomics network server 102 examines the gene definitions from multiple databases to see if they are specifically, but not generally, affected by the drug of interest. In addition, the use of, for example, Google Scholar^TMAnd/or PubMed, evaluating published literature containing text strings containing gene names or precursor gene names or equivalent protein names plus any functions related to the drug of interest. These may include affinities reproducibly discovered to bind the same pharmacodynamic target at the drug of interestBinding affinity studies for molecules bound with an affinity within 10 times the force. Second, the pharmacogenomics web server 102 examines each gene for all mutations that contain SNPs, variable number of tandem repeat motifs, duplications, and all other known mutation alterations, extending as linear sequence +10kb from the transcription start site and stop codon of the gene, as examined in a genome browser (e.g., UCSC genome browser or Ensembl genome browser). If any of these mutations are present in published literature or in sources such as unpublished clinical trial data and are involved in the action of the drug of interest, including efficacy, adverse events or first pass metabolism, the mutation is added to a preliminary gene set comprising a pharmacogenomic network (block 192). Third, particularly for complex tissues such as brain, skin, and cardiovascular system, the pharmacogenomics web server 102 qualitatively performs a consistency mapping to compare the expression of all genes in this final set with the expression (if known) of the drug of interest that plays its role. Genes whose expression does not match the pharmacodynamic substrate of the drug of interest are discarded (block 192). Finally, use is made of, for example, Ingeneity Pathway Analysis^TMThird party software is used to check the connectivity of the gene set (block 193). Using fisher's right-hand side exact test, if the pharmacogenomic network server 102 determines that significant interconnectivity exists based on published literature, the genes are placed into the preliminary gene set of the pharmacogenomic network that includes the drug of interest. Any genes that do not form a linked network are discarded as non-candidate genes for the pharmacogenomic network (block 194).

The pharmacogenomics network server 102 can perform knowledge-based remediation using gene expression patterns where the overlap indicates functional correspondence. This example is shown in figure 38, which may be present in the case where the genes of the ketamine pharmacogenomics network exhibit statistically significant overlap (P < 1E-56; Fisher exact test), where the drug exerts its rapid effect in the human brain, including the Anterior Cingulate Cortex (ACC) and the Frontal Cortex (FC), but no overlap in the somatosensory cortex (SSC), Occipital Cortex (OC) or Corpus Callosum (CC).

Drug pharmacogenomics network reconstruction engine

Figure 11 shows the composition of pharmaco-pharmacogenomic network reconstruction engine 146A using proprietary and public knowledge of the 3D human genome, previously defined TAD, super-enhancers and other properties of the original version of the regulatory genome in human genome construction 19(hg19) or updated sparse human genome construction 38(hg38) as provided in spreadsheet lookup tables. This is a key component of the network reconstruction engine 146A. All experimental data generated from the chromosome conformation capture method, which can be performed in vivo, were subjected to evaluation, generated from causal SNP probing of chromatin datasets in drug-active cell types of interest as shown in fig. 22 and 33, or from public or other private sources. A compact model of drug-induced alterations in the 3D regulatory genome comprising TAD matrix, enhancer-promoter pair, promoter-promoter pair and super-enhancer was developed in 2D or 3D form using SNPs or the above candidate variants identified based on the chosen method, as shown in fig. 6, a combination of 3D modeling comprising human genome architecture in the european miles space (Euclidian space), high resolution optical microscopy using FISH, and/or measures of gene expression (e.g., RNA sequencing, promoter capture Hi-C). After evaluating the pharmacogenomic interaction groups of drugs in chromatin, the resulting pharmacogenomic networks were initially defined. To determine whether network elements are significantly interconnected based on existing biomedical knowledge, third party pathway analysis software is used to provide a significance score. A commonly used program for genetic Pathway Analysis includes Ingenity Pathway Analysis^TMPanther Gene Ontology pathway mapping and KEGG (Kyoto Encyclopedia of Genes and Genomes). In this manner, the network reconstruction engine 146A determines the interconnection between the SNP and the target gene associated with the drug response or adverse event for the particular drug of interest.

Iterative gene set optimization engine

At block 195, the pharmacogenomic network server 102, and more particularly the gene set optimization engine 146C, performs iterative gene set optimization on the identified candidate gene sets for the particular drug of interest in the pharmacogenomic network. An example method for iterative gene set optimization to deconstruct a pharmacogenomic network into subnetworks is shown in the flowchart of fig. 12. Iterative gene set optimization can be performed to identify sub-networks of pharmacogenomic networks. More specifically, iterative Gene set optimization involves the use of an API for all input molecular terms to convert them from, for example, Human Gene Nomenclature Committee (HGNC) names to Gene or long noncoding RNA names. Iterative gene set optimization differs from gene set enrichment methods not only by combining multiple statistical methods, but also by acting like a threshold-dependent method to rank genes in a hierarchical manner, and does not rely on comparison of experimental results, as in the entire distribution test. In contrast, iterative gene set optimization groups genes or long non-coding RNAs using Jaccard Distance (Jaccard Distance) as a ratio of the size of a symmetrically different gene a Δ gene B ═ a ≡ B-a £ B to the union, to measure the similarity between the two genes or long non-coding RNAs first based on the dissimilarity of the terms selected by the user. This can be extended to clusters of related dissimilar gene names. The pharmacogenomics network server 102 then orders these sets into subsets of a subset of clusters of functionally related genes, either automatically using a minimum entropy ordering algorithm such as the COOLCAT algorithm or using a user-defined number of clusters. After optimization using entropy-minimized basis sets, the pharmacogenomic network identification system 100 can employ manual governance to assign efficacy, adverse events, or functional mechanism sub-networks based on known attributes of the mechanism of action of the drug being considered.

Post hoc authentication using third party bioinformatics tools

To scientifically validate the deconstruction of pharmacogenomic networks into sub-networks of mechanisms optimized based on subsets of functional genes, the pharmacogenomic network server 102 retrospectively evaluates each sub-network of pharmacogenomic networks for antecedent terms (molecular function and biological process) of gene ontologies, antecedent terms from drug databases, antecedent canonical pathways determined, e.g., using other proprietary or open-source pathway analysis software, disease risk genetic variant analysis determined, e.g., using other proprietary or open-source pathway analysis software, and determination of upstream xenobiotic modulators using different bioinformatic sources (block 196). Upstream xenobiotic modulators are compared to specific drugs of interest to ensure that the specific drug of interest is the drug most significantly associated with the pharmacogenomic network. More specifically, different bioinformatic sources can be used, ordered according to the respective associations of upstream xenobiotic modulators with the pharmacogenomic network. For example, the upstream xenobiotic modulators whose p-value is lowest relative to the pharmacogenomic network may have the strongest correlation. The pharmacogenomic network server 102 can then determine whether the particular drug of interest is the top-ranked upstream xenobiotic rule or is ranked above a threshold ranking (e.g., top three or top five). In addition, the GWAS directories of the European Bioinformatics Institute (European Bioinformatics Institute), the National Human Genome Research Institute (the National Human Genome Research Institute), and the National Institutes of Health (National Institutes of Health) can be searched to find significant SNP trait associations for each gene in the gene set of each sub-network. By providing examples of statistically significant SNPs from GWAS, it can be provided that mutational impairment of genes contained in each subnetwork provides additional evidence of insight into the normal, undamaged function of the subnetwork. An example method for performing post-hoc validation of a pharmacogenomic network and its constituent sub-networks is shown in the flowchart of figure 13.

In some embodiments, after performing post-hoc validation, the resulting pharmacogenomic networks and component sub-networks for a particular drug of interest are stored, for example, in database 154, as shown in fig. 1A. In some embodiments, the pharmaceutical drug genomic network server 102 may provide the client device 106 with an indication of the pharmacogenomic network and component sub-networks for display to healthcare professionals or researchers 116. The client device 106 can then present the pharmacogenomic network and component sub-networks in a graphical display.

Error correction for pharmacogenomic networks and subnetworks

In some embodiments, the pharmacogenomic network server 102 can tune or tune the pharmacogenomic network and component sub-networks of a particular drug to provide an accurate model for measuring human drug response phenotypes for real-world clinical applications. From studies of population structure using principal component analysis, allele-sharing distance, and other measures, it has been assumed that the distribution of pharmacogenomic phenotypes can be modeled using normal distributions, although there are some outliers. For example, cytochrome P450 gene variation that produces differences in CYP450 isoform activity has previously been considered to be a major determinant of variability in drug response in humans.

One embodiment of the disclosure includes SNPs that enhance regulation of PK gene expression within a subnetwork, as well as other genes located within the same TAD. In a drug and patient dependent context, variations within these networks can affect tissue specific metabolism that extends beyond missense codons. For patients whose TAD boundaries of the PK gene may be compromised, as shown in the example of the PK gene, trans-interaction of the enhancer, which is less constrained by the TAD in which it is located, may lead to drug adverse events.

Fig. 10 shows that the drug metabolism genes and the Human Leukocyte Antigen (HLA) genes involved in immune-related drug adverse responses have strong borders. Of the 13 gene clusters shown here, 12 had the strongest TAD boundary (grade V) in the human genome. These comprise most of genes encoding cytochrome P450 enzymes (CYP genes), glucuronidase (UGT) superfamily genes, Sulfotransferase (SULT) superfamily genes, N-acetyltransferase (NAT) family genes, and HLA genes. Mutations, such as SNPs, located in the TAD borders of these genes have deleterious effects on drug metabolism and drug response variation, including the occurrence of adverse drug events in the human population.

It is also recognized that additional variables play a role in human drug response, including the effects of social conditions and other environmental factors, which are variables that are often difficult to measure.

Figure 17 shows the pharmacogenomic network topology that can be used by the system as a model of action for most psychopharmaceuticals. In fig. 17A, a template model of the Central Nervous System (CNS) contains a drug and a cassette comprising: (1) chromatin remodeling; (2) Efficacy (EFF) and/or Neuroplasticity (NP); (3) CNS Adverse Events (AE); and (4) centrally active pharmacokinetic enzymes (PK) and hormones (H). For some drugs, Systemic Pharmacokinetics (SPK) is a major determinant of variability in human drug response, and peripheral adverse drug events involving the immune system (IAE) are problematic. In fig. 17B, different pharmacogenomic network topologies for psychotropic drugs with examples of different biological profiles fitted to different two-dimensional topologies are shown. Thus, the topology of sub-network types shown in fig. 17 can be used to deconstruct a pharmacogenomic network into constituent sub-networks, where the sub-network types of most psychopharmaceuticals can contain two or more of: (1) chromatin remodeling; (2) Efficacy (EFF) and/or Neuroplasticity (NP); (3) CNS Adverse Events (AE); (4) centrally active pharmacokinetic enzymes (PK) and hormones (H); (5) systemic Pharmacokinetics (SPK); and (6) peripheral adverse drug events involving the immune system (IAE).

Figure 14 illustrates a machine learning-based method 600 by which computationally predicted sub-network metrics of efficacy and adverse events of pharmacogenomics networks are tuned among populations, including training and test sets, to obtain an accurate discretization of response phenotypes. In some embodiments, the pharmacogenomic network server 102 performs the method 600 shown in fig. 14. Likewise, in some embodiments, the method 600 may be implemented in a set of instructions stored on a non-transitory computer readable memory and executable on one or more processors on the pharmacogenomic network server 102. In any case, the method 600 increases the accuracy of the postulated distribution of human response phenotypes for a particular drug developed by computational analysis by training such drug-specific sub-networks using the pharmacogenomic network identification system 100, wherein the human drug response phenotypes are derived from a sub-network of efficacy and adverse events. In this way, the pharmacogenomics network server 102 increases the utility of the distribution of human response phenotypes for use in real-world clinical applications, making it useful for any reference-based comparison metric performed in medicine or life sciences.

Matching pharmacogenomic networks of reference drugs to patients

The learning architecture for training the pattern matching sub-network includes pre-training a reference set (reference numeral 710), as shown in FIG. 15. This is further described with reference to method 700 shown in fig. 15, which may also be performed by pharmacogenomic network server 102. More specifically, at block 704, the pharmacogenomics network server 102 develops a pattern matching sub-network of the patient derived from the patient input biological sample and jointly develops individual trained pattern metrics containing features of the sub-networks of efficacy and adverse events that are combined feature representation metrics (block 712). To determine similarity to the reference set (blocks 706, 708), two different reference-patient metric pairs contain an accurate measure of similarity and an output similarity score for each of efficacy and adverse events (blocks 714, 716). At block 702, a biological sample obtained from a patient, which may be a cheek swab, a blood or urine sample, is subjected to targeted enhancer SNP genotyping as well as combined chromosome conformation capture and RNA-seq. The pharmacogenomics network server 102 then performs the analysis necessary to construct an input patient-specific map of the efficacy and adverse event sub-network for the particular drug of interest at block 704. These patient-specific, drug-induced sub-network patterns can be further processed using bayesian probability calculations to fill in sparse or missing data. When a new patient is entered as input, the pre-trained reference set of sub-networks of drug-specific efficacy and adverse events for pattern matching is optimized again for subsequent patients, resulting in a more accurate measure of pharmacogenomic variability with enhanced clinical utility between humans. The matching task assumes that the patches undergo the same feature encoding before computing and outputting the similarity score, greatly improving efficiency while reducing computational requirements.

Thus, each input set (reference number 710) and patient set (reference number 720)) is constructed differently with feature set extraction and inference on sparse data using probability calculations based on bayesian distributions to increase the accuracy of the reference and patient maps. The trained feature network is based on the "siemese" network approach, the constraint being that the two sets must share the same parameters. When completed, the patient's drug-induced training pattern network is coupled with a network obtained from a reference database, paired efficacy feature set pairs, and adverse event feature set pairs. These provide the basis for the development of trained efficacy metrics that attempt to match all features from the patient and a reference set of drugs of interest, and trained adverse event metrics. These pairwise match scores yield individual efficacy and adverse event similarity scores between the reference and the patient.

A further refinement of this embodiment is to develop a set of reference pattern matches for each patient that can be used to create a patient-specific database of such reference maps and updated in a periodic manner as additional biological samples are obtained from the patient in a longitudinal manner, obtained over time in a clinical setting, or in an outpatient pharmacy.

Method for identifying drug target in silicon

Fig. 16A and 16B illustrate another method 800 for developing molecules that are druggable pharmacodynamic targets using a pharmacogenomics network and a sub-network of efficacy and adverse events for a particular drug of interest. In some embodiments, the pharmacogenomic network server 102 performs the method 800 shown in figure 16A. Likewise, in some embodiments, the method 800 may be implemented in a set of instructions stored on a non-transitory computer readable memory and executable on one or more processors on the pharmacogenomic network server 102.

In any case, a previously unidentified gene encoding a druggable pharmacodynamic target can be linked to a sub-network of efficacy of the pharmacogenomic network of the particular drug, wherein the connectivity to the plurality of genes in the sub-network of adverse events of the pharmacogenomic network of the particular drug is minimal at the level of pharmacogenomic modulation. In the example illustrated in fig. 16A and 16B, a ketamine pharmacogenomics network was used and the druggable target was PPP1R1B (protein phosphatase 1 regulation inhibitor subunit 1B) (reference numeral 804), a bidirectional signal transduction molecule regulated by the neurotransmitter dopamine. PPP1R1B gene 804 is located within the same TAD as NEUROD2 (neuronal differentiation 2) gene 802, and the same enhancers in both neuronal and astrocytic cell lines regulate both genes. In addition, TADs containing these genes interact in trans with TADs containing DRD2 (dopamine receptor D2)806 and ADORA2A (adenosine A2a receptor) genes 808, also under the control of the same neurons and astrocyte enhancers in their respective TADs. Following the methods described herein, the PPP1R1B pathway, although not a known pharmacogenomic network, is significantly interconnected in the human brain (p 1E-88), and seven of these genes are contained in ketamine pharmacogenomic networks, including BDNF, DRD2, GRIA1, GRIN1, GRIN2A and KLF6, and PPP1R 1B. Four genes were contained in the ketamine neuroplasticity subnetwork of the ketamine pharmacogenomics network, as shown in fig. 29 and 16C, while the other four were contained within the glutamate receptor subnetwork of the ketamine pharmacogenomics network. Comparing the location of gene expression within the PPPR1B pathway in different human brain regions of different genes, only 14 were expressed at detectable levels in the human brain. In addition to GRIA1, GRIN1, and GRIN2A, which are more widely expressed in the human brain, the expression of the remaining 11 genes in this pathway as determined by RNA-seq data in humans showed a significantly restricted gene expression pattern, limited to the foretail, nucleus accumbens, and shell (fig. 16D).

At block 810, the pharmacogenomics network server 102 performs bioinformatics analysis on PPP1R1B pathways limited to those 14 genes expressed in the human brain. Bioinformatic analysis showed that the 14 genes were significantly associated with neuronal differentiation, neuronal development and modulation of neurogenesis and CNS development and opioid signaling. These properties are common to the neuroplasticity sub-networks of valproic acid pharmacogenomic networks shown in figures 18 and 21, the neuroplasticity sub-networks of ketamine pharmacogenomic networks shown in figure 32, and the lithium pharmacogenomic networks shown in figure 43.

As shown in figures 39, 40 and 41, the pharmacogenomic network identification system 100 can reveal complementary properties of existing drug products that can be combined together as new drug compounds, can be administered sequentially, or can identify similar combinations of drugs from the same drug class to provide a more comprehensive therapy for a given clinical indication. Figure 41 demonstrates valproic acid neurogenesis pharmacogenomic subnetwork enrichment to stimulate early neurogenesis and ketamine neuroplasticity subnetwork responsible for late neurogenesis. Figure 39 shows that one mechanism by which this combination of therapeutic agents operates is by sequential acetylation and deacetylation of the histone lysine 9(H3K9) moiety. Thus, valproic acid combines histone deacetylation inhibition with induction of the neural progenitor BAF chromatin remodeling complex for transforming multipotent neuronal precursor cells into committed neuronal progenitor cells as shown in figure 39 (top), and ketamine converts committed neuronal progenitor cells into terminally differentiated neurons by activating HUSH (human silencing complex) and H3K9 methylation (figure 39, bottom).

Figure 40 demonstrates the gene set enrichment of nuclear genes contained within valproic acid pharmacogenomics network (figure 40A) and ketamine pharmacogenomics network (figure 40B). Figure 40A demonstrates that analysis of valproic acid pharmacogenomic networks enriched both H3K9 histone deacetylase activity and neurogenesis, and figure 40B shows that ketamine pharmacogenomic networks enriched both H3K9 histone methyltransferase activity and neuronal differentiation.

Thus, as shown in fig. 41, it is possible to combine these approved drugs in clinical use to provide a comprehensive solution for providing both early and mid-to late-stage neurogenesis, including mechanisms of neurogenesis, neuronal proliferation, neuronal differentiation, and synaptic integration. For example, a first drug (e.g., valproic acid) may be administered to a patient at a first time point, and then a second drug (e.g., ketamine) may be administered to the patient at a second time point after the first time point. Thus, this combination of FDA-approved therapeutics can be used not only in disease states where neuronal cell loss is a characteristic feature of the disorder, but also in the aged human brain to maintain gray matter integrity. The disease state may comprise neurological disorders, neurodegenerative disorders such as frontotemporal dementia, alzheimer's disease and parkinson's disease, as well as neuropsychiatric disorders including bipolar disorder and schizophrenia, as well as acute brain injury.

For example, a method for treating a patient suffering from a neurodegenerative disorder can comprise administering valproic acid to the patient and administering ketamine to the patient. In some embodiments, the method may comprise obtaining a biological sample of the patient and comparing or having compared the biological sample to one or more SNPs in the valproic acid pharmacogenomic network that are associated with neurogenesis. The method may further comprise comparing or having compared the biological sample to one or more SNPs in the ketamine pharmacogenomic network that are associated with neuronal differentiation. In response to determining that the biological sample of the patient comprises a SNP associated with neurogenesis in a valproic acid pharmacogenomic network and a SNP associated with neuronal differentiation in a ketamine pharmacogenomic network, valproic acid and ketamine can be administered to the patient to treat a neurogenic disorder in the patient.

More generally, the pharmacogenomic network identification system 100 can identify pharmacogenomic networks for any number of drugs. For the first drug and the second drug, the pharmacogenomic network identification system 100 can then compare the properties (e.g., drug response phenotype) associated with the genes within the pharmacogenomic network and/or component sub-network of the first drug to the properties (e.g., drug response phenotype) associated with the genes within the pharmacogenomic network and/or component sub-network of the second drug to identify complementary properties between the first drug and the second drug. When complementary properties of a drug group (e.g., early neurogenesis and late neurogenesis) are identified, the drug group can be reused to test as a therapeutic agent for a particular disease or disease state.

Figure 18 shows valproic acid pharmacogenomics networks in the human central nervous system and their component sub-networks as the output of the system. The gene subnetwork consists of: (1) chromatin remodeling and H3K9 acetylation; (2) neurogenesis and anti-epileptic, antimanic and anti-migraine properties; (3) an adverse event; and (4) hormone regulation and pharmacokinetics.

Figure 19 shows the results of ex post-facto bioinformatic analysis of valproic acid pharmacogenomic networks and their accompanying network topology models. Figure 19A shows that the most significant disease annotations for the genes contained in the valproic acid pharmacogenomic network are epilepsy or neurodevelopmental disorder, cognitive impairment, mood disorders, migraine and mania. It should be noted that valproic acid is suitable for use in a single and adjunctive therapy for the treatment of simple and complex absence episodes and for adjunctive treatment of mania in bipolar and other mood disorders in patients suffering from multiple absence types including absence episodes, as well as for the prevention and alleviation of migraine. A common adverse event of valproic acid is cognitive cloudiness (cognitive impairment).

Figure 19B shows the most significant drugs that act as upstream modulators of valproic acid pharmacogenomics networks, including valproic acid (p-value 5.20E-114; Fisher exact test), HDAC inhibitor trichostatin a (p-3.21E-35), and nicotine (p-5.57E-21).

Figure 19C shows valproic acid pharmacogenomic network fit model network topology tag 1, where the deconstructed gene cluster networks include Chromatin Remodeling (CR), neuroplasticity and drug efficacy (NP, EFF), adverse events and neurotransmission (AE, NT), and pharmacokinetics and hormone regulation (PK, H). The non-CNS, peripheral System Pharmacokinetic (SPK) sub-networks of valproic acid pharmacogenomics networks cannot be determined from the output of this system.

Figure 28 shows an example of trans-interactions in 3D chromatin space of valproic acid pharmacogenomic networks as an output of the methods and systems described herein. Genome-wide Hi-C data mapping was performed using SNPs as data probes, including SNPs contained within valproic acid subnetworks and obtained from GWAS catalogs, including those significantly associated with disease risk and valproic acid response variation and dissociation. These results were used to detect both cis-and trans-interactions with other members of the valproic acid pharmacogenomic pathway within human neurons. Fig. 28A shows a complete genomic map that is key to understanding the gene-gene interactions shown in fig. 28B-28I as determined by the Hi-C method. Figure 28B shows Hi-C contacts between GABBR1 (a gene encoding the receptor for gamma-aminobutyric acid (GABA), the major inhibitory neurotransmitter in the human CNS, and enhancer mutations in this gene underlying brain disorders such as epilepsy) and CRHR1 and CRHR1-IT1 (genes important for neural progenitor cell differentiation under the control of several super enhancers, corticotropin binding in the adult brain, and mutations associated with anxiety and mania). Figure 28C shows Hi-C contacts between GABRG2 (the gene encoding a mutation in GABA receptors significantly associated with febrile and infantile epilepsy) and KCNJ3 (the gene encoding a mutation in the human CNS and potassium channels significantly associated with cognition and epilepsy) in human neuronal space. Figure 28C shows spatial Hi-C contacts between RUNX1 as a transcription factor and HDAC9 as a member of the histone deacetylase superfamily of human neurons. Mutations in the enhancer and super enhancer of both genes were significantly associated with hair loss and mild hair loss, an adverse event associated with valproic acid therapy. Figure 28E shows spatial Hi-C contacts between GABRB2, GABRG2 (genes encoding GABA receptors and mutated for ataxia and epilepsy related) and KCNQ5 (genes whose mutation of the super enhancer is significantly associated with autosomal mental retardation and intellectual impairment) in human neurons. Figure 28F shows spatial Hi-C contacts between NEUROD1 (the major transcription factor involved in neurogenesis and involved in valproic acid responses in humans) and NEUROG3 (the transcription factor involved in the lineage commitment of neural progenitors to neurons) in human neurons. Figure 28G shows spatial Hi-C contacts between GRIN2A (encoding N-methyl-D-aspartate (NMDA) receptor members whose mutations are significantly associated with schizophrenia, bipolar disorder, and mania) and ANK 3(a gene encoding a synaptic cytoskeletal member and whose mutations are significantly associated with bipolar disorder, sleep pattern, and schizophrenia) in human neurons. Fig. 28H shows spatial Hi-C contacts between GRIN2B (the gene encoding the NMDA receptor member) and SNCA (the gene whose mutation in presynaptic protein and its super-enhancer is significantly associated with late-stage onset parkinson's disease) in human neurons. Figure 28I shows spatial Hi-C contacts between PAX6 (the gene encoding the major transcription factor responsible for early development of the human central nervous system, eyes and nose) and SOX2 (the gene associated with PAX6 that controls the neural progenitor BAF remodeling complex responsible for neurogenesis) in human neurons.

Figure 20 demonstrates a valproic acid pharmacogenomic adverse event sub-network, and post hoc bioinformatics analysis shows that the valproic acid adverse event sub-network is significantly associated with cancer, severe psychological disorders, gastrointestinal disorders, tremor, and hair loss.

Figure 21 demonstrates a valproic acid pharmacogenomic neuroplasticity subnetwork, and a post hoc bioinformatics analysis shows that the glutamate receptor subnetwork is significantly associated with neurogenesis, neuronal differentiation and neuronal proliferation, and disease states including epilepsy, mood disorders and migraine.

Figure 22 shows an example of 237 unique GWAS disease risk and pharmacogenomic SNPs from the efficacy and adverse event profile that can be used to differentiate individual patients' anti-epileptic, anti-manic and anti-migraine properties in response to valproic acid. Figure 22A shows that multiple GWAS disease risk SNPs located within the valproic acid adverse event subnetwork can be annotated as an enhancer and super enhancer associated with reduced efficacy of hair loss, CNS-mediated gastrointestinal disorders, and the mixed antidepressants bupropion in bipolar depression. Figure 22B shows that the valproic acid pharmacogenomic neuroplasticity subnetwork contains multiple GWAS disease risk and pharmacogenomic SNPs that can be annotated as significantly associated with chronic migraine, epilepsy, bipolar disorder (international classification of disease (ICD) code F31.0-F31.64) and super-enhancer and the efficacy of valproic acid in bipolar mania.

Figure 23 illustrates the validation of the system by comparison with the results of the experiments and public data contained in the most widely used open source and commercial drug databases. Venn Diagram (Venn Diagram) is the output of valproic acid pharmacogenomic network genes in human neural tissue of the invention compared to genes differentially regulated by valproic acid following drug administration to control the brain of pigs (ferae), and is considered to be derived from valproic acid using Ingenity Pathway Analysis in all human tissues^TM(IPA; Qiagen; GmBH) and Kyoto encyclopedia of genes (KEGG) and DrugBank (LINCS database from national institute of health, USA) and DrugCentral regulated Limited Gene sets. The gene sets of the valproic acid pharmacogenomic network as the output of the system described herein exhibited the highest degree of overlap in all one-to-one comparisons of results and data from all these sources.

Figure 24 is a list of all genes contained in the Chromatin Remodeling (CR) subnetwork of the valproic acid pharmacogenomic network in the human brain.

Figure 25 is a list of all genes contained in the neuroplasticity and drug efficacy (NP, EFF) subnetwork of the valproic acid pharmacogenomics network in the human brain.

Figure 26 is a list of all genes contained in the adverse events and neurotransmission (AE, NT) subnetwork of the valproic acid pharmacogenomic network in the human brain.

Figure 27 is a list of all genes contained in the pharmacokinetic and hormone regulation (PK, H) sub-network of valproic acid pharmacogenomic network in human brain.

Figure 29 shows the results of post-hoc bioinformatic analysis of ketamine pharmacogenomic networks and their accompanying network topology models. Figure 29A shows the most significant disease annotations for the genes contained in the ketamine pharmacogenomic network are schizophrenia, refractory depression, bipolar disorder, post-operative delirium and post-operative pain.

Figure 29B shows the most significant drugs that act as upstream modulators of the ketamine pharmacogenomics network, including ketamine (p-value 6.26E-33 ketamine; Fisher exact test), morphine (p-1.97E-17), and nicotine (p-6.62E-17).

Figure 29C shows ketamine pharmacogenomic network fit model network topology tag 2, where the deconstructed gene set networks included Chromatin Remodeling (CR), Adverse Events (AE) and Neurotransmission (NT), neuroplasticity and drug efficacy (NP, EFF), and pharmacokinetics and hormone regulation (PK, H). The non-CNS, peripheral System Pharmacokinetic (SPK) sub-networks of the ketamine pharmacogenomic network cannot be determined from the output of this system.

Fig. 37 shows an example of trans-interactions of ketamine pharmacogenomic networks in 3D chromatin space as an output of the methods and systems described herein. Whole genome Hi-C data mapping was performed using SNPs as data probes, including SNPs contained within the ketamine subnetwork and obtained from the GWAS catalogue, including those significantly associated with disease risk and ketamine antidepressant response variation and dissociation. These results validate pathway analysis and demonstrate both cis-and trans-interactions with other members of the ketamine pharmacogenomic pathway within human neurons. These pharmacogenomic contacts are significantly enriched for association with specific super enhancers from cingulate cortex and frontal cortex. Fig. 37A shows a complete genomic map, which is key to understanding the gene-gene interactions shown in fig. 37B-37G. Figure 37B shows Hi-C contact between RASGRF2 (genes associated with synaptic plasticity and alcoholism) and the co-localized nicotinic receptor genes CHRNA3 and CHRNA5 containing SNPs in GWAS that are significantly associated with smoking status. Figure 37C shows the trans-interaction between ROBO2 gene containing multiple SNPs associated with both unipolar depression in GWAS and dissociative and antidepressant responses to ketamine and both the GRIN2B gene and ATF7IP gene. The ATF7IP gene encodes a chromatin remodeling protein essential to the methylation of histone 3 lysine 9(H3K9me3) responsible for HUSH-mediated heterochromatin formation and gene silencing as part of the stable SETDB1 complex. Figure 37D demonstrates pharmacogenomic contacts between TCF4 and the GRM5 gene encoding members of the glutamate metabotropic receptor family and containing an enhancer that is significantly associated with depression in GWAS. Fig. 37E shows how the Hi-C map of the interaction between CACNA1C and GRIN2A and ATF7IP2 genes. In fig. 37F, Hi-C pharmacogenomic contacts obtained from human glutamatergic neurons showed a trans-interaction between CAMK2A gene located on chromosome 5 and GRIN1 and ANAPC2 genes located on chromosome 9. The ANAPC2 protein is part of a complex that controls the formation of synaptic vesicles that cluster at the active region as presynaptic membranes in post-mitotic neurons, and this complex also degrades NEUROD2 as a major component of presynaptic differentiation during neuronal differentiation. Figure 37G shows pharmacogenomic contacts between the DRD2 gene in neurons and the RHOA gene encoding a signaling protein that modulates the cytoskeleton in neurons during synaptic transmission.

Figure 38 demonstrates the overlap between the postmortem human brain and the genes of the ketamine pharmacogenomic network in the region of the human brain where ketamine first exerted its rapid antidepressant response. This provides additional evidence in support of the ketamine pharmacogenomic network and can be demonstrated by comparing neuroanatomical distribution of gene expression data within the ketamine subnetwork with the results of localization of consensus brain maps showing which brain regions were first affected by ketamine obtained from 24 neuroimaging studies. Consensus plots emphasize that the Anterior Cingulate Cortex (ACC), dorsolateral and dorsolateral prefrontal cortex (PFC) and the Supplementary Motor Area (SMA) are always the first human brain area to be activated by the drug. However, other areas of the CNS in neuroimaging studies have been reported to be affected by ketamine in humans immediately after drug administration. These are shown in black in fig. 12A, but do not include the vast majority of brain areas reported as first affected by ketamine in neuroimaging studies examined during the study. As a control, adjacent human brain regions unaffected by ketamine were selected in neuroimaging studies, including the Corpus Callosum (CC), the Occipital Cortex (OC), and the somatosensory cortex (SS). As shown in fig. 12B, genes in the ketamine network were expressed at significantly higher levels in ACC and PFC than in neighboring CCs, SS and OCs where there was no evidence to suggest that ketamine plays a rapid antidepressant role. ACC is part of the cingulate cortex and PFC is part of the frontal cortex.

Figure 30 shows the output of an iterative gene set optimization analysis of ketamine pharmacogenomic networks in the human brain. The larger chloraminoketone pharmacogenomic network contains 3 subnetworks, creating 2 distinct subnetworks. Glutamate receptor networks enrich for synaptic signaling, glutamate receptor signaling, glutamate pathway modulation, and chromatin organization. The pro-xenobiotic (chemical) up-regulator of the glutamate receptor subnetwork is ketamine with p ═ 2.1E-09 (fig. 13A). In contrast, the neuroplasticity subnetwork is enriched for modulation of nervous system development, modulation of neurogenesis, modulation of neuronal differentiation, neurogenesis, and nervous system development (fig. 13B). Neural plasticity subnetworks are shown as being produced by Ingeneity Pathway Analysis^TMSignificant overlap with the "cardiovascular disease, neurological disease and abnormal body injury" network categories identified at p ═ 1E-59, and its pre-xenobiotic up-regulator is also ketamine at p ═ 6E-12.

Figure 31 demonstrates ketamine pharmacogenomic glutamate receptor subnetwork and post hoc bioinformatic analysis shows that glutamate receptor subnetwork is significantly associated with cognitive impairment, bipolar disorder, post-operative delirium, schizophrenia affective disorder, schizophrenia, non-cancerous pain, post-operative pain, emesis, nausea and unconsciousness.

Figure 32 demonstrates ketamine pharmacogenomic neuroplasticity subnetworks, and post hoc bioinformatics analysis shows that neuroplasticity subnetworks are significantly associated with emotional behavior, morphological abnormalities of the nervous system, morphological abnormalities of the brain, depression, anxiety, and morphological abnormalities of neurons.

Figure 33 shows examples of 108 unique GWAS disease risk and pharmacogenomic SNPs from a spectrum of adverse events that can be used to differentiate the efficacy and adverse events profile of individual patients in response to antidepressants as well as other effects of ketamine and its analogs. Figure 33A shows that multiple GWAS disease risk SNPs located within the glutamate receptor subnetwork can be annotated as enhancers related to tobacco smoking status, chronic schizophrenia (ICD diagnostic code F20), and bipolar 1 disorder (ICD diagnostic code F31.0-F31.64). Figure 33B shows that ketamine neuroplasticity sub-network contains multiple GWAS disease risk SNPs that can be annotated as enhancers associated with recurrent depression (ICD code F33), alcoholism, and response to chlorambucil.

Figure 34 is a list of all genes contained in the neuroplasticity and drug efficacy (NP, EFF) subnetwork of ketamine pharmacogenomics network in the human brain.

Figure 35 is a list of all genes contained in Chromatin Remodeling (CR), Adverse Events (AE), and Neurotransmission (NT) subnetworks of the ketamine pharmacogenomic network in the human brain.

Figure 36 is a list of all genes contained in the pharmacokinetic and hormone regulation (PK, H) sub-network of ketamine pharmacogenomic network in the human brain.

Figure 42 shows the results of ex post bioinformatics analysis of lithium pharmacogenomic networks and their accompanying network topology models. Figure 42A shows that the most significant disease notes for the genes contained in the lithium pharmacogenomics network are cognitive impairment, mood disorders, drug-induced tremor, drug-induced weight gain, and schizophrenia.

Figure 42B shows the most significant drugs that act as upstream modulators of lithium pharmacogenomics networks, including lithium chloride (p-value 2.23E-23; Fisher exact test), lithium (p-4.20E-19), and fluoxetine (p-2.94E-14).

Figure 42C shows lithium pharmacogenomic network fit model network topology tag 4, where the deconstructed gene set networks include Chromatin Remodeling (CR), Neuroplasticity (NP), Efficacy (EFF), Adverse Events (AE), and Adverse Events (AE). The non-CNS, peripheral System Pharmacokinetic (SPK) sub-networks of the lithium pharmacogenomic network cannot be determined from the output of this system.

Figure 43 shows high resolution compartmentalization of an example of an embodiment of a gene set sub-network as a detailed output using this system for lithium pharmacogenomics network.

Figure 44 is a list of all genes contained in the Chromatin Remodeling (CR) subnetwork of the lithium pharmacogenomic network in the human brain.

Figure 45 is a list of all genes contained in the Neuroplasticity (NP) subnetwork of the lithium pharmacogenomic network in the human brain.

Figure 46 is a list of all genes contained in the efficacy sub-network (EFF) of the lithium pharmacogenomic network in the human brain.

Figure 47 is a list of all genes contained in the drug-induced tremor, Adverse Event (AE) subnetwork of the lithium pharmacogenomic network in the human brain.

Figure 48 is a list of all genes contained in the drug-induced weight gain, Adverse Event (AE) subnetwork of the lithium pharmacogenomic network in the human brain.

Figure 49 shows the results of ex post bioinformatics analysis of lamotrigine pharmacogenomic networks and their accompanying network topology models. Figure 49A shows the most significant disease annotation for genes contained in lamotrigine pharmacogenomic networks is epilepsy, fibromyalgia, bipolar 1 disorder, mania and treatment resistant schizophrenia.

Figure 49B shows the most significant drugs acting as upstream modulators of lamotrigine pharmacogenomics networks, including lamotriazine (p-value 1.08E-10; Fisher exact test), carbamazepine (p-3.51E-08), and mirtazapine (p-1.06E-07).

Figure 49C shows lamotrigine pharmacogenomic network fit model network topology tag 2, where deconstructed subset of genes include Chromatin Remodeling (CR), Adverse Events (AE) and Neurotransmission (NT), neuroplasticity and drug efficacy (NP, EFF), and pharmacokinetics and hormone regulation (PK, H). The non-CNS, peripheral System Pharmacokinetic (SPK) sub-networks of ketamine pharmacogenomic networks cannot be determined from the output of this system.

Figure 50 demonstrates lamotrigine pharmacogenomic adverse event subnetworks, and post hoc bioinformatic analysis shows that lamotrigine adverse event subnetworks are significantly associated with staffin's syndrome, drug-induced hypersensitivity, progressive cognitive impairment, choreopathic movements, and headache with dizziness.

Figure 51 demonstrates lamotrigine pharmacogenomic neuroplasticity sub-networks, and post hoc bioinformatics analysis shows that lamotrigine pharmacogenomic neuroplasticity sub-networks are significantly associated with neuronal development and neurogenesis as well as disease states including epilepsy, fibromyalgia, bipolar 1 disorder and mania.

Figure 52 is a list of all genes contained in the Chromatin Remodeling (CR) subnetwork of the lamotrigine pharmacogenomic network in the human brain.

Figure 53 is a list of all genes contained in the neuroplasticity and efficacy (NP, EFF) subnetwork of lamotrigine pharmacogenomics network in the human brain.

Figure 54 is a list of all genes contained in Adverse Event (AE) sub-networks of lamotrigine pharmacogenomic networks in the human brain.

Figure 55 is a list of all genes contained in the Pharmacokinetic (PK) sub-network of lamotrigine pharmacogenomic network in human brain.

Figure 56 shows the results of ex post bioinformatics analysis of clozapine pharmacogenomic networks and their accompanying network topology models. Figure 56A shows the most significant disease notes for the genes contained in the clozapine pharmacogenomic network are psychosis, agitation, bipolar spectrum disorders, non-affective psychosis, and treatment resistant schizophrenia.

Figure 56B shows the most significant drugs that act as upstream modulators of the clozapine pharmacogenomics network, including clozapine (p-value 8.85E-110; Fisher exact test), haloperidol (p-1.45E-42) and chlorpromazine (p-6.95E-20).

Figure 56C shows clozapine pharmacogenomic network fit model network topology tag 3, where the deconstructed subset of genes includes Chromatin Remodeling (CR), adverse events, central nervous system (AE), Pharmacokinetics (PK) and adverse events, peripheral immune system (IAE). The non-CNS, peripheral System Pharmacokinetic (SPK) sub-networks of clozapine pharmacogenomic networks cannot be determined from the output of this system.

Figure 57 presents clozapine pharmacogenomic adverse event CNS and peripheral immune system adverse event sub-networks (AE, IAE), and post hoc bioinformatics analysis showed that the clozapine adverse event sub-network was significantly associated with glycometabolism disorders, systemic autoimmune disorders, weight gain, drug-induced neutropenia, and neuronal apoptosis.

Figure 58 demonstrates clozapine pharmacogenomic efficacy sub-network (EFF) and post hoc bioinformatic analysis shows that the chlorazapine efficacy sub-network is significantly associated with psychosis, treatment resistant schizophrenia, non-affective disorder, manic bipolar disorder, recurrent schizophrenia, mania, bipolar disorder, refractory schizophrenia and schizophrenia.

Figure 59 is a list of all genes contained in the Chromatin Remodeling (CR) subnetwork of the clozapine pharmacogenomic network in the human brain.

Figure 60 is a list of all genes contained in the Efficacy (EFF) subnetwork of the clozapine pharmacogenomic network in the human brain.

Figure 61 is a list of all genes contained in the Adverse Events (AE) sub-network of clozapine pharmacogenomic networks in the human brain and in the peripheral immune system.

Figure 62 is a list of all genes contained in the Pharmacokinetic (PK) sub-network of clozapine pharmacogenomic network in the human brain.

Another application of these methods is to identify genes encoding novel druggable molecules whose function is known but not known to be part of the drug efficacy sub-network of the class of drug of interest. As shown in fig. 63, the warfarin pharmacogenomics network contains several candidate drug targets that were not previously known to be members of the anticoagulant pharmacogenomics network for this drug. The method of pharmacogenomic network mapping using SNPs as part of the drug pathway reconstruction enables the addition of the following genes mediating the anticoagulant effect of warfarin as part of sub-network 1: AXL (AXL receptor tyrosine kinase), F9 (coagulation factor IX), merk (MER proto-oncogene, tyrosine kinase), PDGFB (platelet-derived growth factor subunit B), PROC (protein C, inactivators of coagulation factors Va and VIIIa), PROCR (protein C receptor), PROS1 (protein S), and pro Z (protein Z, vitamin K dependent). Some of these novel genes may encode products that are druggable for use as anticoagulant agents.

As shown in figure 63, the identified pharmacogenomic network of warfarin comprises one or more of: ABO, alpha 1-3-N-acetylgalactosamine transferase and alpha 1-3-galactosyltransferase (ABO) gene, aldone reductase family 1 member C3(AKR1C3) gene, AXL receptor tyrosine kinase (AXL) gene, complement factor H-related 5(CFHR5) gene, cytochrome P450 family 2 subfamily C member 19(CYP2C19) gene, cytochrome P450 family 2 subfamily C member 8(CYP2C8) gene, cytochrome P450 family 2 subfamily member C9 (CYP2C9) gene, cytochrome P450 family 3 subfamily A member 4(CYP3A4) gene, cytochrome P450 family 4 subfamily F member 2 (CYP4F2) gene, Erythropoietin (EPO) gene, factor V (F5) gene, factor VII (F7) gene, factor IX (F9) gene, factor X (F10) gene, Blood coagulation factor XI (F11) gene, blood coagulation factor XII (F12) gene, blood coagulation factor XIII A chain (F13A1) gene, fibrinogen alpha chain (FGA) gene, fibrinogen gamma chain (FGG) gene, growth arrest-specific 6(GAS6) gene, histidine-rich glycoprotein (HRG) gene, kininogen 1(KNG1) gene, Lysozyme (LYZ) gene, MER protooncogene, tyrosine kinase (MERK) gene, Matrix Gla Protein (MGP) gene, serum mucoid 1(ORM1) gene, multi-family comb finger 3(PCGF3) gene, platelet-derived growth factor subunit B (PDGFB) gene, protein C, blood coagulation factor Va and VIIIa inactivator (PROC) gene, protein C receptor (PROCR) gene, protein S (PROS1) gene, protein Z, vitamin K-dependent plasma glycoprotein (PROZ) gene, serine Protease (PRSS) 25 gene (8) protease gene, Serine protease 53(PRSS53) gene, sphingosine kinase 1(SPHK1) gene, signal transducer and activator of transcription 3(STAT3) gene, syntaxin 4(STX4) gene, excess 4(SURF4) gene, transient receptor potential cation channel subfamily C member 4-related protein (TRPC4AP) gene, ubiquitin-specific peptidase 7(USP7) gene, vitamin K epoxide reductase complex subunit 1(VKORC 1) gene, vitamin K epoxide reductase complex subunit 1-like 1(VKORC1L1) gene, or Von Willebrand Factor (VWF) gene.

Throughout the specification, multiple instances may implement a component, an operation, or a structure described as a single instance. Although individual operations of one or more methods are illustrated and described as separate operations, one or more of the individual operations may be performed concurrently and nothing requires that the operations be performed in the order illustrated. Structures and functionality presented as separate components in the example configurations may be implemented as a combined structure or component. Similarly, structures and functionality presented as a single component may be implemented as separate components. These and other variations, modifications, additions, and improvements fall within the scope of the subject matter herein.

In addition, certain embodiments are described herein as comprising logic or a plurality of routines, subroutines, applications, or instructions. These may constitute software (e.g., code embodied on a machine-readable medium or in a transmission signal) or hardware. In hardware, an instance, etc. is a tangible unit capable of performing certain operations and may be configured or arranged in a certain manner. In an example embodiment, one or more computer systems (e.g., a stand-alone client or server computer system) or one or more hardware modules (e.g., a processor or a set of processors) of a computer system may be configured by software (e.g., an application or application portion) as a hardware module that operates to perform certain operations described herein.

In various embodiments, the hardware modules may be implemented mechanically or electronically. For example, a hardware module may comprise dedicated circuitry or logic that is permanently configured to perform certain operations (e.g., a special-purpose processor such as a Field Programmable Gate Array (FPGA) or an Application Specific Integrated Circuit (ASIC)). A hardware module may also comprise programmable logic or circuitry (e.g., as embodied in a special purpose processor or other programmable processor) that is temporarily configured by software to perform certain operations. It should be appreciated that the decision to mechanically implement a hardware module in a dedicated and permanently configured circuit or in a temporarily configured circuit (e.g., configured by software) may be driven by cost and time considerations.

Thus, the term "hardware module" should be understood to encompass a tangible entity, as that entity is physically constructed, permanently configured (e.g., hardwired), or temporarily configured (e.g., programmed) to operate in a certain manner or to perform certain operations described herein. In view of embodiments in which the hardware modules are temporarily configured (e.g., programmed), each hardware module need not be configured or instantiated at any one time. For example, where the hardware modules include a general purpose processor configured using software, the general purpose processor may be configured at different times as respective different hardware modules. Thus, software may configure a processor, for example, to constitute a particular hardware module at one time and to constitute a different hardware module at a different time.

A hardware module may provide information to, or receive information from, other hardware modules. Thus, the hardware modules may be considered to be communicatively coupled. In the case of a plurality of such hardware modules being present at the same time, communication may be achieved by signal transmission (e.g. by means of appropriate circuits and buses) connecting the hardware modules. In embodiments in which multiple hardware modules are configured or instantiated at different times, communication between such hardware modules may be accomplished, for example, by storing and retrieving information in a memory structure accessible to the multiple hardware modules. For example, a hardware module may perform operations and store the output of such operations in a memory device to which it is communicatively coupled. Another hardware module may then access this memory device at a later time to retrieve and process the stored output. The hardware modules may also initiate communication with input or output devices and may operate on resources (e.g., sets of information).

Various operations of the example methods described herein may be performed, at least in part, by one or more processors that are temporarily configured (e.g., via software) or permanently configured to perform the relevant operations. Whether temporarily or permanently configured, such processors may constitute processor-implemented modules that operate to perform one or more operations or functions. In some example embodiments, the modules referred to herein may comprise processor-implemented modules.

Similarly, the methods or routines described herein may be implemented at least in part by a processor. For example, at least some of the operations of a method may be performed by one or more processors or processor-implemented hardware modules. The performance of certain operations may be distributed among one or more processors, not only residing within a single machine, but also being deployable across multiple machines. In some example embodiments, the processor or processors may be located at a single location (e.g., within a home environment, an office environment, or as a server farm), while in other embodiments, the processor may be distributed across multiple locations.

The performance of certain operations may be distributed among one or more processors, not only residing within a single machine, but also being deployable across multiple machines. In some example embodiments, one or more processors or processor-implemented modules may be located at a single geographic location (e.g., within a home environment, office environment, or server farm). In other example embodiments, one or more processors or processor-implemented modules may be distributed across multiple geographic locations.

Unless specifically stated otherwise, discussions herein using terms such as "processing," "computing," "calculating," "determining," "presenting," "displaying," or the like, may refer to the action or processes of a machine (e.g., a computer) that manipulates or transforms data represented as physical (e.g., electronic, magnetic, or optical) quantities within one or more memories (e.g., volatile memory, non-volatile memory, or a combination thereof), registers, or other machine components that receive, store, transmit, or display information.

As used herein, any reference to "one embodiment" or "an embodiment" means that a particular element, feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment. The appearances of the phrase "in one embodiment" in various places in the specification are not necessarily all referring to the same embodiment.

Some embodiments may be described using the expression "coupled" and "connected" along with their derivatives. For example, some embodiments may be described using the term "coupled" to indicate that two or more elements are in direct physical or electrical contact. The term "coupled," however, may also mean that two or more elements are not in direct contact with each other, but yet still co-operate or interact with each other. The embodiments are not limited to these ranges.

As used herein, the terms "comprising," "including," "containing," "having," or any other variation thereof, are intended to cover a non-exclusive inclusion. For example, a process, method, article, or apparatus that comprises a list of elements is not necessarily limited to only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Further, unless expressly stated to the contrary, "or" refers to an inclusive or and not to an exclusive or. For example, either of the following satisfies condition a or B: a is true (or present) and B is false (or not present), a is false (or not present) and B is true (or present), and both a and B are true (or present).

In addition, "a" or "an" are used to describe elements and components of embodiments herein. This is merely for convenience and to give a general description. This description and the claims which follow should be understood to include one or at least one and the singular also includes the plural unless it is obvious that it is meant otherwise.

This detailed description is to be construed as merely providing examples, and does not describe every possible embodiment since describing every possible embodiment would be impractical, if not impossible. Many alternative embodiments may be implemented using either current technology or technology developed after the filing date of this application.