阅读说明：本技术 三元杂交培育方法与训练方法及设备 (Ternary hybridization cultivation method and training method and equipment ) 是由 刘宝祥 于 2019-12-05 设计创作，主要内容包括：一种三元杂交培育方法与训练方法及设备。该三元杂交培育方法,包括：第一选育经第一对象和第二对象两元杂交的第一子代,其中,所述第一对象被第一功能基因定位,所述第二对象被第二功能基因定位；第二选育经第一对象和第三对象两元杂交的第二子代,其中,所述第三对象被第三功能基因定位；利用所述第一子代和第二子代繁育目标对象。(A ternary hybridization cultivation method, a training method and equipment. The ternary hybridization cultivation method comprises the following steps: first breeding a first progeny comprising a binary cross of a first object and a second object, wherein said first object is located by a first functional gene and said second object is located by a second functional gene; second breeding a second progeny that is two-way crossed by the first object and a third object, wherein the third object is located by a third functional gene; and breeding the target object by using the first filial generation and the second filial generation.)

1. A ternary hybridization breeding method comprises the following steps:

first breeding a first progeny comprising a binary cross of a first object and a second object, wherein said first object is located by a first functional gene and said second object is located by a second functional gene;

second breeding a second progeny that is two-way crossed by the first object and a third object, wherein the third object is located by a third functional gene;

and breeding the target object by using the first filial generation and the second filial generation.

2. The method of claim 1, wherein the first functional gene maps a first economic trait comprising at least a meat quality trait and a reproductive trait; the second functional gene locates a second economic trait, the second economic trait including at least a growth trait and a reproductive trait; the third functional gene maps a third economic trait, which includes at least a growth trait and a reproductive trait.

3. The method of claim 2, wherein first selecting the first progeny that is bidden by the first object and the second object comprises:

and a first breeding step of stably inheriting first filial generations of the first functional gene and the second functional gene, wherein if the filial generations of the first object and the second object in the binary hybridization reach a fourth economic trait, the filial generations are the first filial generations stably inheriting the first functional gene and the second functional gene, and the fourth economic trait at least comprises a meat quality trait, a propagation trait and a growth trait.

4. The method of claim 3, wherein first breeding the first progeny that stably inherit the first functional gene and the second functional gene comprises:

and if the growth character of the hybrid filial generation reaches 16 months, the weight reaches 600 kilograms, the meat quality character reaches the high-end part, the meat has obvious marbling, and the reproductive character is the fourth economic character of which the fertilization rate reaches a preset threshold, the hybrid is the first filial generation of the first breeding.

5. The method of claim 2, wherein second breeding the second progeny that is binary crossed by the first object and the third object comprises:

and a second breeding step of stably inheriting a second filial generation of the first functional gene and a third functional gene, wherein if the filial generation of the first object and the third object in the binary hybridization reaches a fifth economic trait, the filial generation is the second filial generation stably inheriting the first functional gene and the third functional gene, and the fifth economic trait at least comprises a meat quality trait, a propagation trait and a growth trait.

6. The method of claim 5, wherein the second breeding of the second progeny stably inheriting the first functional gene and the third functional gene comprises:

and if the growth character of the hybrid filial generation reaches 16 months, the weight reaches 600 kilograms, the meat quality character is the fifth economic character of the high-end meat with obvious marbling, and the reproductive character is the fifth economic character of which the fertilization rate reaches a preset threshold value, the hybrid is the second filial generation of the second breeding.

7. The method of claim 2, wherein breeding a target object using the first and second descendants comprises:

and thirdly, breeding a third filial generation which is crossed by the first filial generation and the second filial generation, and fixing the cross breeding target object by using the third filial generation.

8. The method of claim 7, wherein the third breeding of the third progeny that is crossed by the first progeny and the second progeny comprises:

and thirdly, stably breeding a third filial generation which inherits the first functional gene, the second functional gene and the third functional gene, wherein if the filial generation hybridized by the first filial generation and the second filial generation reaches a sixth economic trait, the filial generation is the third filial generation which stably inherits the first functional gene, the second functional gene and the third functional gene, and the sixth economic trait at least comprises a meat quality trait, a propagation trait and a growth trait.

9. The method of claim 2, wherein the method further comprises:

determining a first subject based on the trained first neural network model, wherein the first neural network model determines a first subject located by a first functional gene based on first economic trait data;

determining a second object based on the trained second neural network model, wherein the second neural network model determines a second object located by a second functional gene based on second economic trait data;

determining a third object based on the trained third neural network model, wherein the third neural network model determines a third object that is located by a third functional gene based on third economic trait data.

10. The method according to claims 1-9, wherein the first subject is a Japanese female parent, the second subject is an Angus male parent, and the third subject is a Haford male parent.

Technical Field

The embodiment of the disclosure relates to a ternary hybridization cultivation method, a training method and equipment.

Background

In recent years, with the improvement of living standard, people have more and more demands on beef, and the quality and the taste of the beef also have more and more attention. Not only is the beef required to have juicy and soft mouthfeel and special fragrance, but also the high-end part of the beef is required to have bright marbling on the cross section, and the high-end quality beef with the quality and mouthfeel is mainly imported and is expensive at present.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments of the present invention will be briefly described below, and it is obvious that the drawings described below are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

In the drawings:

FIG. 1 is a flow chart of a method of breeding a ternary cross according to an embodiment of the present disclosure;

FIG. 2 is a process diagram of a convolutional neural network according to an embodiment of the present disclosure;

fig. 3 is a schematic diagram of a training apparatus according to an embodiment of the present disclosure.

Detailed Description

In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

In order to obtain high-end quality beef, a new variety is obtained by mainly utilizing binary hybridization (binary hybridization is also called simple hybridization, namely, two varieties or strains are hybridized to obtain a first generation hybrid) of an exotic cow (male parent) and a local cow (female parent). The exotic cattle can be German cattle, Simmental cattle, Limuzan, Pimont, Haford or Red Angel. The local cattle is local cattle. Although the new variety obtained by binary or multiple hybridization of the foreign cattle and the local cattle has good growth tendency and meat production performance, the grade of the meat at the high-grade part has a certain difference compared with the imported high-grade meat, which is mainly reflected in fat precipitation between muscles, in addition, the new variety has lower reproduction rate, and the feeding cost of the offspring in the fattening period is high.

At least one embodiment of the present disclosure provides a ternary hybridization breeding method, as shown in fig. 1, which is a flow chart of the ternary hybridization breeding method, and the ternary hybridization breeding method at least solves the problems of low fertilization rate of cows hybridized with exotic cows, small offspring physique and poor rough feeding resistance. The ternary hybridization cultivation method comprises the following steps:

s101, first breeding a first filial generation which is subjected to binary crossing of a first object and a second object, wherein the first object is positioned by a first functional gene, and the second object is positioned by a second functional gene;

step S102, second breeding a second filial generation which is subjected to binary crossing of the first object and a third object, wherein the third object is positioned by a third functional gene;

and S103, breeding the target object by using the first filial generation and the second filial generation.

Here, the first subject is a female cow. Specifically, the first object is pure blood Japanese black hair and female cattle, the weight of an adult cow is about 620 kg, the weight of a bull is about 950 kg, and a calf is fattened by 27 months, so that the weight reaches over 700 kg, and the average daily gain is over 1.2 kg. Japan and cattle are the best quality beef cattle recognized in the world today, and the meat has obvious marbling, also called as snowflake meat. Since the meat of Japan and cattle is juicy and tender, the saturated fatty acid content in the muscle fat is very low, the flavor is unique, the meat value is extremely high, the meat is regarded as 'national treasure' in Japan, and the meat is extremely expensive in the Western Europe market. Japan and cattle are very precious high-quality beef cattle variety resources in Japan.

Here, the second object is the angses father. Angses have good performance in meat and are considered to be one of the typical species of specialized beef cattle in the world. It is also called black horn-free black cattle because it is an important feature of black fur and horn-free. The cattle has low trunk, is firm, has small and square head, wide forehead, deep trunk, cylindrical shape, short and straight limbs, wide front and back steps, full muscle and typical body type of modern beef cattle. The carcass quality is high, the meat yield is high, the slaughter rate is generally 60-65%, and the muscle marbling is good. The average live weight of an Angus bull adult bull is 700-900 kg, the average cow is 500-600 kg, the average birth weight of a calf is 25-32 kg, the average adult height bull and cow are 130.8 cm and 118.9 cm respectively, the daily weight gain in lactation period is 900-1000 g, and the average daily weight gain in fattening period (within 1.5 years) is 0.7-0.9 kg.

Here, the third object is the heford male parent. The Haifute cattle, which is produced in Haifute county in England south of England, is the oldest early-maturing small and medium-sized beef cattle variety in the world, and is a special beef cattle variety in America, namely horn-free Hafu cattle, after years of improvement and breeding in America, the Hafu cattle has the advantages of wide body, developed chest, full muscles of the whole body, short head, wide forehead, short and thick neck, developed neck sag and front and back regions, straight and wide back and waist, expanded ribs, straight and short limbs, cylindrical trunk and typical cuboid shape of beef cattle. The hair of the quilt is reddish brown except for white at the head, neck, abdomen, lower parts of limbs and tail end, and the skin is orange red. The calves are born heavy, the male is 34 kg, and the female is 32 kg; the weight of the body reaches 400 kilograms when the body is 12 months old, and the average daily gain is more than 1 kilogram. The weight of the adult is 1100 jin for male cattle at 1000-. When slaughtering is carried out 400 days after birth, the slaughtering rate is 60-65%, and the net meat rate reaches 57%. The meat is tender, the taste is delicious, the deposited fat among muscle fibers is rich, and the meat is in a marble shape. The Haifute cattle has the characteristics of strong physique, coarse feeding resistance, suitability for grazing, high meat yield and the like.

In step S101, a first offspring is first bred which is a binary cross of a first object and a second object, wherein the first object is located by a first functional gene and the second object is located by a second functional gene. The first functional gene locates a first economic trait, and the first economic trait at least comprises a meat quality trait and a propagation trait; the second functional gene maps a second economic trait, which includes at least a growth trait and a reproductive trait. Optionally, the first economic trait may further comprise a growth trait and an appearance trait, and the second economic trait may further comprise a meat quality trait and an appearance trait.

In addition, with reference to molecular breeding techniques, the economic traits of animals (such as the above-mentioned economic traits, milk production, carcass yield, and intramuscular fat deposition) belong to the category of quantitative traits, and are controlled by multiple alleles and show co-dominance. First breeding a first progeny that is binary crossed by a first object and a second object comprises: and a first breeding step of stably inheriting a first functional gene and a first filial generation of a second functional gene, wherein if the filial generation of the first object and the second object in the binary hybridization reaches a fourth economic trait, the filial generation is the first filial generation stably inheriting the first functional gene and the second functional gene, and the fourth economic trait at least comprises a meat quality trait, a reproductive trait and a growth trait. For example, if the hybrid progeny has a growth trait of 16 months and a weight of 600 kg, a meat quality trait of a high-end part of meat has obvious marbling, and a reproductive trait of a fourth economic trait in which the fertilization rate reaches a preset threshold, the hybrid progeny is the first progeny of the first breeding. The specific breeding standard is as follows:

growth performance: the weight of the patient reaches 600 kg in 16 months;

the net meat rate: 40-42%, or more than 40%;

the meat at the high end has obvious marbling;

the meat quality of the high-end part is tender and succulent, and reaches the production standard of western cuisine.

Fertilization rate: greater than a preset threshold.

In step S102, a second progeny is selected that is two-way crossed by the first object and a third object, wherein the third object is located by a third functional gene. The third functional gene maps a third economic trait, which includes at least a growth trait and a reproductive trait. Here, the second breeding of the second progeny that is two-way crossed by the first object and the third object includes: and a second breeding step of stably inheriting a second filial generation of the first functional gene and a third functional gene, wherein if the filial generation of the first object and the third object in the binary hybridization reaches a fifth economic trait, the filial generation is the second filial generation stably inheriting the first functional gene and the third functional gene, and the fifth economic trait at least comprises a meat quality trait, a propagation trait and a growth trait.

Optionally, if the hybrid progeny after the binary crossing of the first object and the third object reaches the fifth economic trait that the growth trait is 16 months, the body weight reaches 600 kg, the meat quality trait is the high-end part of meat with obvious marbling, and the reproductive trait is the fertilization rate reaches the preset threshold, the hybrid progeny is the second progeny of the second breeding.

In step S103, the target object is bred using the first child and the second child. The breeding of the target object by using the first filial generation and the second filial generation comprises the following steps: and thirdly, breeding a third filial generation which is crossed by the first filial generation and the second filial generation, and fixing the cross breeding target object by using the third filial generation. Here, the third breeding stably inherits the first functional gene, the second functional gene and the third progeny of the third functional gene, wherein if the hybrid progeny of the cross of the first progeny and the second progeny reaches a sixth economic trait, the breeding progeny is the third progeny stably inheriting the first functional gene, the second functional gene and the third functional gene, and the sixth economic trait includes at least a meat quality trait, a reproductive trait and a growth trait.

The method utilizes the characteristics of high quality meat quality of Hefu cattle and excellent growth characteristics of Hafu cattle and Angus, takes the Hefu cattle as a female parent and the Hefu cattle and Angus as male parents respectively, performs respective hybridization to obtain hybrid filial generations, and performs breeding on the hybrid filial generations through the breeding standard to obtain a first filial generation and a second filial generation. For example, the first generation is male line, the second generation is female line, and the superior first generation and the second generation are crossed to produce the third generation, so that the third generation can obtain the superior performance of beef cattle (and cattle, Haeford and Angel) with three qualities. The third filial generation is fixed in a crossing way to establish a new strain core group (embryo transplantation breeding technology can also be utilized to carry out rapid propagation), and the filial generation generated by the third filial generation in the crossing way can have good production performance and high-quality meat quality and completely has the standard and the special quality of high-end beef cattle. In addition, the offspring population generated by the cross-crossing and fixing of the third offspring has strong propagation capacity and coarse feeding resistance, and the feeding cost is economical and practical.

It should be noted that the breeding direction of the first filial generation, the second filial generation and the third filial generation can be reversely deduced through market application, demand standard and economic value.

In the prior art, the first subject, the second subject and the third subject are mainly determined by observing dominant traits of cattle. In the present disclosure, further comprising: determining a first object based on the trained first neural network model, wherein the first neural network model determines a first object located by the first functional gene based on the first economic trait data; determining a second object based on the trained second neural network model, wherein the second neural network model determines the second object located by the second functional gene based on the second economic trait data; determining a third object based on the trained third neural network model, wherein the third neural network model determines a third object located by a third functional gene based on the third economic trait data.

The neural network model may be any suitable type of neural network model and may be trained, parameterized, in advance using a large amount of data before being used to determine the first object or the second object or the third object. Next, the neural network will be described by taking a convolutional neural network as an example. Convolutional Neural Networks (CNN) are locally connected networks. After acquiring the first economic trait data, the second economic trait data or the third economic trait data of the cattle to be selected, the convolutional neural network sequentially passes through a plurality of processing processes (such as each hierarchy in fig. 2) and then outputs an identification result, wherein the identification result can be positioned by the first functional gene, the second functional gene or the third functional gene. The processing procedure of each level may include: convolution (convolution) and downsampling (down-sampling). Features of a given first economic trait data, or second economic trait data, or third economic trait data may be extracted by performing a convolution with the first economic trait data or second economic trait data by one convolution kernel, and different characteristics may be extracted by different convolution kernels. In general terms, the calculation method of the convolutional layer can be performed according to the following formula:

where σ represents an activation function; imgMat represents an economic trait matrix; w represents a convolution kernel; "

"denotes a convolution operation; b represents an offset value. The processing procedure of each level may further include a normalization process (e.g., LCN, localconstant normalization) and the like as necessary.

The embodiment of the disclosure can perform convolution operation on the first economic trait data, the second economic trait data or the third economic trait data of the cattle to be selected through a convolution neural network, for example, to determine a first object, a second object or a third object, wherein the determined first object is strongly positioned by the first functional gene, and has excellent quality and meat quality characteristics; the determined second object is strongly positioned by the second functional gene and has excellent growth characteristics; the determined third object is strongly positioned by the third functional gene and has excellent growth characteristics.

It should be noted that the steps illustrated in the flowchart of fig. 2 may be performed in a computer system such as a set of computer-executable instructions and that, although a logical order is illustrated in the flowchart, in some cases, the steps illustrated or described may be performed in an order different than presented herein. Through the above description of the embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but the former is a better implementation mode in many cases. Based on such understanding, the technical solutions of the present disclosure may be embodied in the form of a software product, which is stored in a storage medium (e.g., ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal device (e.g., a mobile phone, a computer, a server, or a network device) to execute the method of the embodiments of the present disclosure.

At least one embodiment of the present disclosure also provides a training device, which may have multiple implementation manners, for example, the training device may be implemented by a single computer, may be implemented by multiple computers, may be implemented by being deployed in a cloud, or the like, or may be implemented by a combination of these manners. As shown in fig. 3, the training device comprises a processor 301 and a memory 302, the memory 302 being configured to store computer program instructions adapted to be loaded by the processor and to perform the above-described method of training a first neural network model, or training a second neural network model, or training a third neural network model. The processor 301 may be any suitable processor, such as a central processing unit, a microprocessor, an embedded processor, and the like, and may adopt an architecture such as X86, ARM, and the like; the memory 302 may be a variety of suitable storage devices including, but not limited to, magnetic storage devices, semiconductor storage devices, optical storage devices, etc., and may be arranged as a single storage device, an array of storage devices, or a distributed storage device, which are not limited by embodiments of the present disclosure.

At least one embodiment of the present disclosure also provides a computer-readable non-transitory storage medium storing computer program instructions that, when executed by a computer, perform the above-described method of training a first neural network model, or training a second neural network model, or training a third neural network model.

It should be noted that, for the sake of simplicity, the above-mentioned embodiments of the training device and the storage medium are all described as a series of acts or a combination of modules, but those skilled in the art should understand that the present disclosure is not limited by the described sequence of acts or the connection of modules, because some steps may be performed in other sequences or simultaneously and some modules may be performed in other connection manners according to the present disclosure.

Those skilled in the art should also appreciate that the embodiments described in this specification are all one embodiment, and the above-described embodiment numbers are merely for description, and the acts and modules involved are not necessarily essential to the disclosure. In the above embodiments of the present disclosure, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.

The above is merely an embodiment of the present disclosure, and it should be noted that, for those skilled in the art, several modifications and decorations can be made without departing from the principle of the present disclosure, and these modifications and decorations should also be regarded as the protection scope of the present disclosure.