Application of Genomic Data for Reliability Improvement of Pig Breeding Value Estimates

Correlations and Variations Between the Major Biochemical Parameters of the Blood of Hybrid Swine

In modern animal husbandry, increasing attention is given to mathematical modeling and statistical methods, especially for evaluating commercial hybrids. Our aim was to evaluate the phenotypic and genetic variability of biochemical parameters of blood serum of the 56 hybrid boars (Large White × Landrace × Duroc) raised in feeding stations (Russia) through mathematical modeling. The particular variances and covariances of traits were calculated using the limited maximum likelihood model and the REMLF90 programs. A narrow range of variability was found for major biochemical parameters in relationship with the "FFG-factor" ("fattening period × final live weight × gain"), including the majority of the metabolites (p ≤ 0.05). The highest values of the genetic correlations were observed for the "total protein" parameter with albumins (0.78), globulins (0.94), creatinine (0.99), and enzymes: AST (0.98), ALT (0.80), etc. Phenotypic and genetic relationships showed fairly high correlation coefficients (0.5-0.8). It is important to emphasize that most of the studied amino acids (alanine, arginine, aspartic acid and asparagine, glutamic acid and glutamine, glycine, isoleucine, leucine, serine, threonine, tyrosine, valine) were significantly associated with the "FFG-factor" (p ≤ 0.05). The proposed approach provides reliable data on metabolite variability and correlations.

PNNGS, a multi-convolutional parallel neural network for genomic selection

Genomic selection (GS) can accomplish breeding faster than phenotypic selection. Improving prediction accuracy is the key to promoting GS. To improve the GS prediction accuracy and stability, we introduce parallel convolution to deep learning for GS and call it a parallel neural network for genomic selection (PNNGS). In PNNGS, information passes through convolutions of different kernel sizes in parallel. The convolutions in each branch are connected with residuals. Four different Lp loss functions train PNNGS. Through experiments, the optimal number of parallel paths for rice, sunflower, wheat, and maize is found to be 4, 6, 4, and 3, respectively. Phenotype prediction is performed on 24 cases through ridge-regression best linear unbiased prediction (RRBLUP), random forests (RF), support vector regression (SVR), deep neural network genomic prediction (DNNGP), and PNNGS. Serial DNNGP and parallel PNNGS outperform the other three algorithms. On average, PNNGS prediction accuracy is 0.031 larger than DNNGP prediction accuracy, indicating that parallelism can improve the GS model. Plants are divided into clusters through principal component analysis (PCA) and K-means clustering algorithms. The sample sizes of different clusters vary greatly, indicating that this is unbalanced data. Through stratified sampling, the prediction stability and accuracy of PNNGS are improved. When the training samples are reduced in small clusters, the prediction accuracy of PNNGS decreases significantly. Increasing the sample size of small clusters is critical to improving the prediction accuracy of GS.

Incorporating the pedigree information in multi-environment trial analyses for improving common vetch

Common vetch is one of the most profitable forage legumes due to its versatility in end-use which includes grain, hay, green manure, and silage. Furthermore, common vetch is one of the best crops to rotate with cereals as it can increase soil fertility which results in higher yield in cereal crops. The National Vetch Breeding Program located in South Australia is focused on developing new vetch varieties with higher grain and dry matter yields, better resistance to major diseases, and wider adaptability to Australian cropping environments. As part of this program, a study was conducted with 35 field trials from 2015 to 2021 in South Australia, Western Australia, Victoria, and New South Wales with the objective of determining the best parents for future crosses and the vetch lines with highest commercial value in terms of grain yield production. A total of 392 varieties were evaluated. The individual field trials were combined in a multi-environment trial data, where each trial is identified as an environment. Multiplicative mixed models were used to analyze the data and a factor analytic approach to model the genetic by environment interaction effects. The pedigree of the lines was then assembled and incorporated into the analysis. This approach allowed to partition the total effects into additive and non-additive components. The total and additive genetic effects were inspected across and within environments for broad and specific selections of the lines with the best commercial value and the best parents. Summary measures of overall performance and stability were used to aid with selection of parents. To the best of our knowledge, this is the first study which used the pedigree information to breed common vetch. In this paper, the application of this statistical methodology has been successfully implemented with the inclusion of the pedigree improving the fit of the models to the data with most of the total genetic variation explained by the additive heritable component. The results of this study have shown the importance of including the pedigree information for common vetch breeding programs and have improved the ability of breeders to select superior commercial lines and parents.

Case Study on Increasing Breeding Value Estimation Reliability of Reproductive Traits in Serbian Highly Prolific Large White and Landrace Sows

Simple Summary

In the Republic of Serbia, pig selection in recent decades has been based on genetic improvement of growth and carcass quality traits. Genetic improvement of reproductive traits of pigs was based on the so-called phenotypic selection. The introduction of modern information systems and the availability larger dataset have opened the possibility to perform genetic estimation of reproductive traits within the main breeding programme of the Republic of Serbia. Using the methods of gene flow and connectedness evaluation, our study investigated the possibility of improving the reliability of estimating the breeding value of reproductive traits in highly productive sows. We believe that these methods could lead to a systematic improvement of the genetic value of reproductive traits in sows. Thus far, none of the methods for estimating the degree of connectedness between herds in pigs has been used in the preparation of the National Breeding Programme of the Republic of Serbia.

Abstract

This study investigated the influence of the degree of connectedness on the reliability of the estimated breeding values (EBVs). The focal trait in the study was the number of piglets born alive (NBA) from sows of the highly prolific Large White and Landrace sows. An analysis included total of 58,043 farrowing’s during the 2008–2020 period. BLUP procedure was used to estimate the breeding values for NBA for the three herds separately and after merging all three herds into one herd. The model for EBV estimation included the following fixed factors: parity, genotype, seasons, litter sire, herds, sow age at farrowing, weaning-conception interval, length of previous lactation, and the following random effects: common litter environment, permanent litter environment, and direct additive genetic effect of animal. Heritability values for NBA ranged from 0.048 to 0.097, depending on the data included in the analysis. The connectedness between herds was analysed using the connectedness rating (CR) and the gene flow (GF) methods. CR among the observed herds ranged from 0.245 to 0.994%, depending on the data included. The exchange of genetic material between all three herds was determined using GF method. The high degree of connectedness determined by the CR and GF method had a strong effect on EBV reliability. The average EBV reliability ranged from 0.520 to 0.867, depending on the data included. The increase in average reliability was observed in both cases when the data were added, both in the analysis of average reliability for purebred animals and when crossbreeds were added, where an increase in this value was also observed. The increase in average EBV reliability is a consequence of the greater amount of information included in the joint evaluation. In conclusion, we believe that our research will improve EBV reliability and help in further selection work in the Republic of Serbia.

Weighted Single-Step Genomic Best Linear Unbiased Prediction Method Application for Assessing Pigs on Meat Productivity and Reproduction Traits

Changes in the accuracy of the genomic estimates obtained by the ssGBLUP and wssGBLUP methods were evaluated using different reference groups. The weighting procedure’s reasonableness of application Pwas considered to improve the accuracy of genomic predictions for meat, fattening and reproduction traits in pigs. Six reference groups were formed to assess the genomic data quantity impact on the accuracy of predicted values (groups of genotyped animals). The datasets included 62,927 records of meat and fattening productivity (fat thickness over 6–7 ribs (BF1, mm)), muscle depth (MD, mm) and precocity up to 100 kg (age, days) and 16,070 observations of reproductive qualities (the number of all born piglets (TNB) and the number of live-born piglets (NBA), according to the results of the first farrowing). The wssGBLUP method has an advantage over ssGBLUP in terms of estimation reliability. When using a small reference group, the difference in the accuracy of ssGBLUP over BLUP AM is from −1.9 to +7.3 percent points, while for wssGBLUP, the change in accuracy varies from +18.2 to +87.3 percent points. Furthermore, the superiority of the wssGBLUP is also maintained for the largest group of genotyped animals: from +4.7 to +15.9 percent points for ssGBLUP and from +21.1 to +90.5 percent points for wssGBLUP. However, for all analyzed traits, the number of markers explaining 5% of genetic variability varied from 71 to 108, and the number of such SNPs varied depending on the size of the reference group (79–88 for BF1, 72–81 for MD, 71–108 for age). The results of the genetic variation distribution have the greatest similarity between groups of about 1000 and about 1500 individuals. Thus, the size of the reference group of more than 1000 individuals gives more stable results for the estimation based on the wssGBLUP method, while using the reference group of 500 individuals can lead to distorted results of GEBV.

Genome-Wide Association Studies, Runs of Homozygosity Analysis, and Copy Number Variation Detection to Identify Reproduction-Related Genes in Bama Xiang Pigs

Litter size and teat number are economically important traits in the porcine industry. However, the genetic mechanisms influencing these traits remain unknown. In this study, we analyzed the genetic basis of litter size and teat number in Bama Xiang pigs and evaluated the genomic inbreeding coefficients of this breed. We conducted a genome-wide association study to identify runs of homozygosity (ROH), and copy number variation (CNV) using the novel Illumina PorcineSNP50 BeadChip array in Bama Xiang pigs and annotated the related genes in significant single nucleotide polymorphisms and common copy number variation region (CCNVR). We calculated the ROH-based genomic inbreeding coefficients (FROH) and the Spearman coefficient between FROH and reproduction traits. We completed a mixed linear model association analysis to identify the effect of high-frequency copy number variation (HCNVR; over 5%) on Bama Xiang pig reproductive traits using TASSEL software. Across eight chromosomes, we identified 29 significant single nucleotide polymorphisms, and 12 genes were considered important candidates for litter-size traits based on their vital roles in sperm structure, spermatogenesis, sperm function, ovarian or follicular function, and male/female infertility. We identified 9,322 ROHs; the litter-size traits had a significant negative correlation to FROH. A total of 3,317 CNVs, 24 CCNVR, and 50 HCNVR were identified using cnvPartition and PennCNV. Eleven genes related to reproduction were identified in CCNVRs, including seven genes related to the testis and sperm function in CCNVR1 (chr1 from 311585283 to 315307620). Two candidate genes (NEURL1 and SH3PXD2A) related to reproduction traits were identified in HCNVR34. The result suggests that these genes may improve the litter size of Bama Xiang by marker-assisted selection. However, attention should be paid to deter inbreeding in Bama Xiang pigs to conserve their genetic diversity.