Enhancing the accuracy of genomic prediction is a key goal in genomic selection (GS) research. Integrating prior biological information into GS methods using appropriate models can improve prediction accuracy for complex traits. Genome-wide association study (GWAS) is widely utilized to identify potential candidate loci associated with complex traits in livestock and poultry, offering essential genomic insights. In this study, a GWAS was conducted on 685 Duroc × Landrace × Yorkshire (DLY) pigs to extract significant single-nucleotide polymorphisms (SNPs) as genomic features. We compared two GS models, genomic best linear unbiased prediction (GBLUP) and genomic feature BLUP (GFBLUP), by using imputed whole-genome sequencing (WGS) data on 651 Yorkshire pigs. The results revealed that the GBLUP model achieved prediction accuracies of 0.499 for backfat thickness (BFT) and 0.423 for loin muscle area (LMA). By applying the GFBLUP model with GWAS-based SNP preselection, the average prediction accuracies for BFT and LMA traits reached 0.491 and 0.440, respectively. Specifically, the GFBLUP model displayed a 4.8% enhancement in predicting LMA compared to the GBLUP model. These findings suggest that, in certain scenarios, the GFBLUP model may offer superior genomic prediction accuracy when compared to the GBLUP model, underscoring the potential value of incorporating genomic features to refine GS models.

Human mediated intra‐continental exchange of genetic material among domesticated organisms has never been restricted to a single direction. The introduction of pig breeds between China and Europe aims to enhance economically important traits in local populations. However, the reciprocal introgression pattern, specifically the role of introgressed genes and structural variations (SVs), remains underexplored. A global collection of whole‐genome resequencing data is utilized from 418 pigs to generate comprehensive dataset, including single‐nucleotide polymorphisms (SNPs) as well as SVs. The analysis reveals incomplete linkage disequilibrium between SVs and adjacent SNPs, highlighting the limitations of conventional SNP‐based analyses in capturing the genetic effects of SVs. By examining both population‐level SNPs and SVs, bidirectional introgression between Chinese and European pig populations is characterized. It is identified 3558 bidirectional introgressed genomic segments and 30 SVs, with haplotypes at BMP2 , which are associated with improved body size. The origin and allele frequency trajectory of the BMP2 segment are further validated using ancient genomes, suggesting that the body size‐enhancing haplotype likely originated from ancient European populations and has since maintained a relatively high allele frequency. Overall, the results highlight the significant role of bidirectional introgression in shaping the genetic composition and phenotypic traits in modern pig breeds.

During the process of pork production, the carcasses of pigs are divided and sold, which provides better economic benefits and market competitiveness for pork production than selling the carcass as a whole. Due to the significant cost of post-slaughter phenotypic measurement, the genetic architecture of tenderloin weight (TLNW) and rib weight (RIBW)—important components of pig carcass economic value—remain unknown. In this study, we conducted genome-wide association studies (GWAS) for TLNW and RIBW traits in a population of 431 Duroc × Landrace × Yorkshire (DLY) pigs. In our study, the most significant single nucleotide polymorphism (SNP) associated with TLNW was identified as ASGA0085853 (3.28 Mb) on Sus scrofa chromosome 12 (SSC12), while for RIBW, it was Affx-1115046258 (172.45 Mb) on SSC13. Through haplotype block analysis, we discovered a novel quantitative trait locus (QTL) associated with TLNW, spanning a 5 kb region on SSC12, and a novel RIBW-associated QTL spanning 1.42 Mb on SSC13. Furthermore, we hypothesized that three candidate genes, TIMP2 and EML1, and SMN1, are associated with TLNW and RIBW, respectively. Our research not only addresses the knowledge gap regarding TLNW, but also serves as a valuable reference for studying RIBW. The identified SNP loci strongly associated with TLNW and RIBW may prove useful for marker-assisted selection in pig breeding programs.

Genomic prediction (GP), which uses genome-wide markers to estimate breeding values, is a crucial tool for accelerating genetic progress in livestock and plant breeding. The accuracy of GP depends on several factors, including the statistical model, marker density, and cross-validation strategy. This study evaluated these factors to optimize GP accuracy for eight economically important carcass and body traits in a Duroc × (Landrace × Yorkshire) (DLY) pig population. This study used 50 K SNP chip data from 1494 DLY pigs, which were imputed to the whole genome sequence (WGS) level. Seven different models were compared, including GBLUP, ssGBLUP, and five Bayesian models. The ssGBLUP model consistently outperformed other models across all traits, with prediction accuracies ranging from 0.371 to 0.502. Further analyses showed that prediction accuracy improved with increasing cross-validation folds and marker density, particularly in the low-density panel. However, the improvement plateaued in medium-to-high-density scenarios. These findings underscore the importance of carefully selecting the model, marker density, and cross-validation strategy to optimize GP accuracy for carcass and body traits in commercial pigs. The insights from this study can guide breeders and researchers in maximizing genetic progress in pig breeding programs.

The Duroc × (Landrace × Yorkshire) (DLY) pig is a cornerstone of three-way crossbreeding system. Nevertheless, advances in commercial crossbred performance have been constrained by the dearth of high-resolution genomic resources for this key population. Here, we report the sequencing and assembly of 16 haplotype-resolved, chromosome-level genome assemblies derived from eight DLY pigs. These assemblies exhibited high continuity (contig N50: 18.17–29.54 Mb) and completeness (BUSCO: 99.3–99.4%), with sequences successfully localized to the 19 chromosomes. Genome annotation revealed an average of 21,922 protein-coding genes and 44.66% repetitive sequences per assembly. Comparative genomic analysis against the current reference genome Sscrofa11.1 enabled the construction of a non-redundant SV catalog comprising 130,416 variants, nearly half of which (48.99%) were novel relative to existing pig pan-genome SV panel. These SVs clustered non-randomly into 231 “SV hotspots” that were significantly enriched in protein-coding genes and putative regulatory elements. Functional analyses further linked these SV hotspots to quantitative trait loci (QTLs) associated with economically important traits. A focused analysis of a 3.43 Mb hotspot on chromosome 1, overlapping a known QTL for average daily gain, revealed eight high-frequency SVs in open chromatin regions near candidate genes (NCS1, HMCN2, FUBP3, ABL1, and FIBCD1), suggesting a cis-regulatory mechanism that may influence gene expression. Collectively, this work provides the first haplotype-resolved genomic resource for commercial crossbred pigs, and establishes a foundational framework for deciphering the genomic architecture of hybrid vigor and advancing precision breeding in swine.

The wild boar (Sus scrofa) is a widely distributed native wildlife species in China, where it plays an important role in diverse ecosystems. Current research on wild boars primarily focused on their geographical distribution, epidemiological investigation, and genetic relationship with domesticated pigs. However, our understanding of the genomic characteristics of wild boar remains limited. Here, we present two high-quality chromosome-level genome assemblies for Chinese wild boars, generated using PacBio HiFi and Hi-C sequencing technologies. The genome assemblies have a mean size of 2.62 Gb, with an average contig N50 of 58.0 Mb and a scaffold N50 of 140.2 Mb. Both genomes were successfully anchored to 18 autosomes and the X chromosome. The Benchmarking Universal Single-Copy Orthologs (BUSCO) assessment showed a high level of genome completeness at 96.3%. On average, a total of 25,851 protein-coding genes were annotated, with a transcriptome BUSCO completeness score of 94.4%. We identified 1.24 Gb of repeat sequences, accounting for 47.31% of the genome. Among these, retroelements represent 38.26%, with LINEs being the most prevalent. The two high-quality genome assemblies presented in this study provide valuable genomic resources for future research and will support the conservation and management of wild boar genetic diversity.