Heng Du

The increasing availability of multi-omics data is promising in enhancing genomic prediction in breeding and human genetics. However, integrating multi-omi

Background: Multibreed genomic prediction (MBGP) is crucial for improving prediction accuracy for breeds with small populations, for which limited data are often available. Recent studies have demonstrated that partitioning the genome into nonoverlapping blocks to model heterogeneous genetic (co)variance in multitrait models can achieve higher joint prediction accuracy. However, the block partitioning method, a key factor influencing model performance, has not been extensively explored. Results: We introduce mbBayesABLD, a novel Bayesian MBGP model that partitions each chromosome into nonoverlapping blocks on the basis of linkage disequilibrium (LD) patterns. In this model, marker effects within each block are assumed to follow normal distributions with block-specific parameters. We employ simulated data as well as empirical datasets from pigs and beans to assess genomic prediction accuracy across different models using cross-validation. The results demonstrate that mbBayesABLD significantly outperforms conventional MBGP models, such as GBLUP and BayesR. For the meat marbling score trait in pigs, compared with GBLUP, which does not account for heterogeneous genetic (co)variance, mbBayesABLD improves the prediction accuracy for the small-population breed Landrace by 15.6%. Furthermore, our findings indicate that a moderate level of similarity in LD patterns between breeds (with an average correlation of 0.6) is sufficient to improve the prediction accuracy of the target breed. Conclusions: This study presents a novel LD block-based approach for multibreed genomic prediction. Our work provides a practical tool for livestock breeding programs and offers new insights into leveraging genetic diversity across breeds for improved genomic prediction.

Infectious diseases are among the primary constraints to pig production, and the globalization of the pig industry has contributed to the emergence and spread of pathogens. However, there is a lack of comprehensive genomic surveillance on the Eurasian scale, resulting in the prevalence and evolution of pig pathogenic viruses and bacteria are still unknown. In this study, we proposed a protocol to identify viral and bacterial sequences and estimate the abundance accurately based on the whole-genome sequencing data of the blood samples. Through whole-genome analysis of 685 Eurasian pigs, we constructed the blood virome and bacteriome landscape. There were a total of 15 pathogenic bacteria, 12 pathogenic viruses, and porcine endogenous retrovirus were identified. We divided 685 Eurasian pigs into three subgroups and discovered significant differences in the viral and bacterial composition, prevalence, and abundance among subgroups. Besides, we performed the quantitative Polymerase Chain Reaction experiment to quantify the copy number of porcine endogenous retrovirus and confirm the reliability of the proposed protocol. Furthermore, we constructed the phylogenetic tree of porcine parvovirus 6 and the results suggested that large-scale transportation across China provides viral connectivity between geographically distinct localities, potentially facilitating the spread of viruses. We also discovered the ADAM28 and ADAMDEC1 genes that may relate to porcine lymphotropic herpesvirus, and the ATF4 gene that may correlate with porcine cytomegalovirus. Our study provides new insights into the genomic investigation and epidemiology of viruses and bacteria, in turn helping to prevent viral and bacterial infectious diseases in pigs.

Nonparametric models have recently been receiving increased attention due to their effectiveness in genomic prediction for complex traits. However, regular nonparametric models cannot effectively differentiate the relative importance of various SNPs, which significantly impedes the further application of these methods for genomic prediction. To enhance the fitting ability of nonparametric models and improve genomic prediction accuracy, a weighted kernel ridge regression model (WKRR) was proposed in this study. For this new method, different weights were assigned to different SNPs according to the p-values from GWAS, and then a KRR model based on these weighted SNPs was constructed for genomic prediction. Cross-validation was further adopted to choose appropriate hyper-parameters during the weighting and prediction process for generalization. We compared the predictive accuracy of WKRR with the genomic best linear unbiased prediction (GBLUP), BayesR, and unweighted KRR using both simulated and real datasets. The results showed that WKRR outperformed unweighted KRR in all simulated scenarios. Additionally, WKRR achieved an average improvement of 1.70% in accuracies across all traits in a mice dataset and 2.17% for three lactation-related traits in a cattle dataset compared to GBLUP, and yielded competitive results compared to BayesR. These findings demonstrated the great potential of weighted nonparametric models for genomic prediction.

Performing joint genomic predictions for multiple breeds (MBGP) to expand the reference size is a promising strategy for improving the prediction for limited population sizes or phenotypic records for a single breed. This study proposes an MBGP model—mbBayesAB, which treats the same traits of different breeds as potentially genetically related but different, and divides chromosomes into independent blocks to fit heterogeneous genetic (co)variances. Best practices of random effect (co)variance matrix priors in mbBayesAB were analyzed, and the prediction accuracies of mbBayesAB were compared with within-breed (WBGP) and other commonly used MBGP models. The results showed that assigning an inverse Wishart prior to the random effect and obtaining information on the scale of the inverse Wishart prior from the phenotype enabled mbBayesAB to achieve the highest accuracy. When combining two cattle breeds (Limousin and Angus) in reference, mbBayesAB achieved higher accuracy than the WBGP model for two weight traits. For the marbling score trait in pigs, MBGP of the Yorkshire and Landrace breeds led to a 6.27% increase in accuracy for Yorkshire validation using mbBayesAB compared to that using the WBGP model. Therefore, considering heterogeneous genetic (co)variance in MBGP is advantageous. However, determining appropriate priors for (co)variance and hyperparameters is crucial for MBGP.

Background: As warthogs (Phacochoerus africanus) have innate immunity against African swine fever (ASF), it is critical to understand the evolutionary novelty of warthogs to explain their specific ASF resistance. Methods: Here, we present two completed new genomes of one warthog and one Kenyan domestic pig as fundamental genomic references to elucidate the genetic mechanisms of ASF tolerance. Results: Multiple genomic variations, including gene losses, independent contraction, and the expansion of specific gene families, likely molded the warthog genome to adapt to the environment. Importantly, the analysis of the presence and absence of genomic sequences revealed that the DNA sequence of the warthog genome had an absence of the gene lactate dehydrogenase B (LDHB) on chromosome 2 compared with the reference genome. The overexpression and siRNA of LDHB inhibited the replication of the African swine fever virus. Combined with large-scale sequencing data from 42 pigs worldwide, the contraction and expansion of tripartite motif-containing (TRIM) gene families revealed that TRIM family genes in the warthog genome are potentially responsible for its tolerance to ASF. Conclusion: Our results will help improve the understanding of genetic resistance to ASF in pigs.