阅读说明：本技术 一种宏病毒组流程的优化分析方法 (Optimized analysis method for macrovirome process ) 是由 薛正晟 李鸿毅 寇文伯 孙子奎 于 2020-12-29 设计创作，主要内容包括：本发明公开了一种宏病毒组流程的优化分析方法,1)序列质控,2)序列拼接,3)病毒序列预测,4)病毒序列聚类处理,5)物种注释,本发明分析数据更为准确,能拼接出长度较长的contig,从而提高比对结果的准确性和注释结果的可靠性。(The invention discloses an optimized analysis method of a macrovirus group process, which comprises the steps of 1) sequence quality control, 2) sequence splicing, 3) virus sequence prediction, 4) virus sequence clustering processing and 5) species annotation.)

1. An optimization analysis method for a macrovirome process is characterized by comprising the following steps:

1) performing sequence quality control, namely performing quality filtration on an original sequence by using fastp software to obtain a high-quality sequence;

2) splicing the sequences, namely splicing the high-quality sequences obtained in the last step by using a megahit software package to obtain spliced contig files;

3) forecasting a virus sequence, namely forecasting the virus sequence of the contig file obtained in the last step by using virporter software to obtain the contig sequence which is forecasted to be the virus;

4) performing virus sequence clustering treatment, namely selecting the contig sequences predicted to be viruses in the step 3), and performing clustering analysis on the sequences by using vcontact2 software to obtain clustered result files;

5) and (4) annotating species, and calculating the annotation result of the virus sequence by using the clustering result obtained in the last step and using an algorithm of lca.

2. The method of claim 1, wherein the method comprises: identifying potential linker sequences at the 3' end by using Cutadapt as raw data in the step 1), and truncating at the identified linker sequences; the matching length with the adaptor sequence is more than or equal to 3bp, and the base mismatching rate is allowed to be less than or equal to 20 percent.

3. The optimized analysis method for the macrovirome process as claimed in claim 1, wherein the fast software p in step 1) adopts a sliding window method to perform quality screening on the sequences: the window size is 5bp, the movement is started from the first base position of the 5 'end, the average quality of the base in the window is required to be more than or equal to Q20, and the sequence is cut off from the 3' end base of the window with the first average quality value lower than Q20; removing the sequence with the sequence length less than 50bp after the quality screening; after the quality screening, sequences containing fuzzy bases in the sequences are removed.

4. The method of claim 1, wherein the method comprises: the contig length in the step 3) is more than 1 kb; virsorter takes out category1, category2, category4 and category 5; the rest contigs larger than 5kb are extracted to obtain the protein sequence of the contigs as vHMM.

5. The method of claim 4, wherein the protein sequence of contigs is vHMM as follows:

one contigs requires at least five proteins HMM-annotated hit, accounting for more than 10%, with less than 20% KO-annotated genes and less than 40% PFAM-annotated genes;

the number of HMM hit-containing genes is not less than the number of PFAM hit genes;

HMM hit is greater than or equal to 60% of the total base factor.

6. The method for optimizing analysis of macrovirome flow as claimed in claim 1, wherein in step 5), vContact2 software is used to perform species annotation on contig sequences predicted to be viruses.

Technical Field

The invention relates to the technical field of gene detection, in particular to an optimization analysis method for a macrovirome process.

Background

The macrovirome sequencing refers to high-throughput sequencing of all viral genomes of a sample species, and analysis of the composition of a viral population or the composition and functions of genes in a specific environment through analysis of the composition of sequencing sequences. By means of the analysis of the difference of the virus constitution in different environments, the relation between the virus and the host can be analyzed, and a marker virus or a gene with a specific function can be searched. To achieve some of the objectives. The most common method currently used for macrovirome analysis is based on reads analysis. The principle of the method is that the reads are compared with a virus database, and the sequence on the comparison is annotated as the virus. However, because the reliability and accuracy of the comparison of reads are not high, the obtained result is often questioned, and therefore, a data analysis method based on splicing needs to be introduced.

The existing reads-based analysis method has the following defects:

(1) reads is too short, often only 150bp, so that the comparison is carried out, and the obtained result is often false positive;

(2) the results of the reads-based analysis method often contain a large number of virus results, and most of the results are not credible.

Disclosure of Invention

The invention provides an optimization analysis method of a macrovirome process.

The scheme of the invention is as follows:

an optimized analysis method for a macrovirome process comprises the following steps:

1) performing sequence quality control, namely performing quality filtration on an original sequence by using fastp software to obtain a high-quality sequence;

2) splicing the sequences, namely splicing the high-quality sequences obtained in the last step by using a megahit software package to obtain spliced contig files;

3) forecasting a virus sequence, namely forecasting the virus sequence of the contig file obtained in the last step by using virporter software to obtain the contig sequence which is forecasted to be the virus;

4) performing virus sequence clustering treatment, namely selecting the contig sequences predicted to be viruses in the step 3), and performing clustering analysis on the sequences by using vcontact2 software to obtain clustered result files;

5) and (4) annotating species, and calculating the annotation result of the virus sequence by using the clustering result obtained in the last step and using an algorithm of lca.

Preferably, the raw data in step 1) is to identify a potential linker sequence at the 3' end by using Cutadapt, and truncate at the identified linker sequence; the matching length with the adaptor sequence is more than or equal to 3bp, and the base mismatching rate is allowed to be less than or equal to 20 percent.

As a preferred technical scheme, the fast software p in the step 1) adopts a sliding window method to perform quality screening on the sequences: the window size is 5bp, the movement is started from the first base position of the 5 'end, the average quality of the base in the window is required to be more than or equal to Q20, and the sequence is cut off from the 3' end base of the window with the first average quality value lower than Q20; removing the sequence with the sequence length less than 50bp after the quality screening; after the quality screening, sequences containing fuzzy bases in the sequences are removed.

As a preferred technical solution, the contig in the step 3) is longer than 1 kb; virsorter takes out category1, category2, category4 and category 5; the rest contigs larger than 5kb are extracted to obtain the protein sequence of the contigs as vHMM.

As a preferred technical scheme, the protein sequence of the contigs is made into vHMM as follows:

one contigs requires at least five proteins HMM-annotated hit, accounting for more than 10%, with less than 20% KO-annotated genes and less than 40% PFAM-annotated genes;

the number of HMM hit-containing genes is not less than the number of PFAM hit genes;

HMM hit is greater than or equal to 60% of the total base factor.

Preferably, in step 5), the species annotation is performed on the contig sequences predicted to be viruses by using vContact2 software.

The invention provides an optimized analysis method of a macrovirome process, which comprises the following steps of 1) sequence quality control, wherein fastp software is used for carrying out quality filtration on an original sequence to obtain a high-quality sequence; 2) splicing the sequences, namely splicing the high-quality sequences obtained in the last step by using a megahit software package to obtain spliced contig files; 3) forecasting a virus sequence, namely forecasting the virus sequence of the contig file obtained in the last step by using virporter software to obtain the contig sequence which is forecasted to be the virus; 4) performing virus sequence clustering treatment, namely selecting the contig sequences predicted to be viruses in the step 3), and performing clustering analysis on the sequences by using vcontact2 software to obtain clustered result files; 5) and (4) annotating species, and calculating the annotation result of the virus sequence by using the clustering result obtained in the last step and using an algorithm of lca.

The invention has the advantages that: the analysis data is more accurate compared with a method based on reads, and contigs with longer length can be spliced, so that the accuracy of comparison results and the reliability of annotation results are improved;

the virus contigs are clustered by using the vcontact2 software, and are annotated based on the lca algorithm, so that the result is reliable, the problem of false positive is avoided, and the data analysis is more accurate.

Drawings

FIG. 1 is a schematic flow chart of an embodiment of the present invention.

Detailed Description

In order to make up for the above deficiencies, the present invention provides an optimized analysis method for a macrovirus process to solve the above problems in the background art.

An optimized analysis method for a macrovirome process comprises the following steps:

1) performing sequence quality control, namely performing quality filtration on an original sequence by using fastp software to obtain a high-quality sequence;

2) splicing the sequences, namely splicing the high-quality sequences obtained in the last step by using a megahit software package to obtain spliced contig files;

3) forecasting a virus sequence, namely forecasting the virus sequence of the contig file obtained in the last step by using virporter software to obtain the contig sequence which is forecasted to be the virus;

4) performing virus sequence clustering treatment, namely selecting the contig sequences predicted to be viruses in the step 3), and performing clustering analysis on the sequences by using vcontact2 software to obtain clustered result files;

5) and (4) annotating species, and calculating the annotation result of the virus sequence by using the clustering result obtained in the last step and using an algorithm of lca.

Identifying potential linker sequences at the 3' end by using Cutadapt as raw data in the step 1), and truncating at the identified linker sequences; the matching length with the adaptor sequence is more than or equal to 3bp, and the base mismatching rate is allowed to be less than or equal to 20 percent.

The fast software p in the step 1) adopts a sliding window method to carry out quality screening on the sequence: the window size is 5bp, the movement is started from the first base position of the 5 'end, the average quality of the base in the window is required to be more than or equal to Q20, and the sequence is cut off from the 3' end base of the window with the first average quality value lower than Q20; removing the sequence with the sequence length less than 50bp after the quality screening; after the quality screening, sequences containing fuzzy bases in the sequences are removed.

The contig length in the step 3) is more than 1 kb; virsorter takes out category1, category2, category4 and category 5; the rest contigs larger than 5kb are extracted to obtain the protein sequence of the contigs as vHMM.

The protein sequence of the contigs was written as vmmm:

one contigs requires at least five proteins HMM-annotated hit, accounting for more than 10%, with less than 20% KO-annotated genes and less than 40% PFAM-annotated genes;

the number of HMM hit-containing genes is not less than the number of PFAM hit genes;

HMM hit is greater than or equal to 60% of the total base factor.

The species annotation of contig sequences predicted to be viruses was performed in step 5) using the vContact2 software.

In order to make the technical means, the creation characteristics, the achievement purposes and the effects of the invention easy to understand, the invention is further described with the specific embodiments.

Example (b):

in step S101, sequence quality control is performed, and quality control processing is performed on an existing original sequence to obtain a high-quality sequence.

In the present embodiment, Cutadapt is used to identify potential linker sequences at the 3' end (few occurrences will occur) and truncate at the identified linker sequences. It is required that the matching length with the linker sequence (R1: AGATCGGAAGAGCACACGTCTGAACTCCAGTCA; R2: AGATCGGAAGAGCGTCGTGTAGGGAAAGAGTGT) is at least 3bp and that a base mismatch ratio of at most 20% is allowed.

After removing the 3' end connector sequence, using fastp to carry out quality screening on the sequence by adopting a sliding window method: the window size is 5bp, the window moves from the first base position of the 5 'end, the average base quality in the window is required to be more than or equal to Q20 (namely the average base sequencing accuracy is more than 99 percent), and the sequence is cut off from the 3' end base of the window with the first average quality value lower than Q20.

After the quality screening, the sequence with the sequence length less than 50bp is removed.

After the quality screening, sequences containing fuzzy bases in the sequences are removed.

In step S102, megahit is used to splice the high quality energy storage obtained in the previous step, so as to obtain a spliced contig file.

In the embodiment of the invention, firstly, the parameter setting of K-mer [27,127] is used, megahit (https:// hku-bal.github.io/megabox /) is called, the two-end sequence of each sample is respectively assembled and spliced from the beginning (De novo), a Contigs sequence is constructed through a De Bruijn diagram, and the assembly and splicing effect evaluation is carried out on the generated Contigs sequence.

In step S103, the contig obtained in the previous step is subjected to prediction of a virus sequence using virporter software, and a contig sequence predicted as a virus is obtained.

In the embodiment of the invention, Virsorter is firstly used for predicting contigs (only the length is more than 1 kb; Virsorter updates a PFAM database to be the latest), and the category1, 2, 4 and 5 in the contigs are taken out (according to the literature of ' Unswitching Earth's video ' and the like); extracting the protein sequence of the rest contigs (requiring more than 5kb) to be used as a vHMM (downloading an HMM model according to the literature Unswitching Earth's virtual), wherein the requirements are as follows:

namely: 1. one contigs requires at least five proteins HMM-annotated hit, and accounts for more than 10%, with less than 20% of the genes annotated with KO and less than 40% of the genes annotated with PFAM.

2. The number of HMM hit-containing genes was not less than the number of PFAM hit-containing genes. (in parallel with the above)

HMM hit greater than or equal to 60% of the total basis factor. (in parallel with the above).

In step S104, a contig sequence predicted as a virus is selected, and the sequence is subjected to clustering analysis using vcontact2 software, so as to obtain a clustering result file.

In step S105, the clustering result obtained in the previous step is used to calculate the annotation result of the virus sequence by using the algorithm lca.

In the embodiment of the invention, vContact2 software is used to perform species annotation on the contig sequences predicted as viruses, because each target sequence may be clustered with a plurality of reference sequences, and the matched reference sequences belong to different classification units, in order to make the analysis strict and reliable and not lose biological significance, we adopt a "Lowest Common Ancestor (LCA)" algorithm to differentiate the reference sequences into the last-stage Common classification before branching of different species as the species classification annotation information of the target sequences.

The foregoing shows and describes the general principles, essential features, and advantages of the invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, which are described in the specification and illustrated only to illustrate the principle of the present invention, but that various changes and modifications may be made therein without departing from the spirit and scope of the present invention, which fall within the scope of the invention as claimed. The scope of the invention is defined by the appended claims and equivalents thereof.