High-quality whole-genome resequencing in large-scale pig populations with pedigree structure and multiple breeds would enable accurate construction of haplotype and robust selection-signature detection. Here, we sequence 740 pigs, combine with 149 of our previously published resequencing data, retrieve 207 resequencing datasets, and form a panel of worldwide distributed wild boars, aboriginal and highly selected pigs with pedigree structures, amounting to 1096 genomes from 43 breeds. Combining with their haplotype-informative reads and pedigree structure, we accurately construct a panel of 1874 haploid genomes with 41,964,356 genetic variants. We further demonstrate its valuable applications in GWAS by identifying five novel loci for intramuscular fat content, and in genomic selection by increasing the accuracy of estimated breeding value by 36.7%. In evolutionary selection, we detect MUC13 gene under a long-term balancing selection, as well as NPR3 gene under positive selection for pig stature. Our study provides abundant genomic variations for robust selection-signature detection and accurate haplotypes for deciphering complex traits in pigs.

对具有谱系结构和多个品种的大规模猪群进行高质量的全基因组重测序将能够准确构建单倍型和稳健的选择特征检测。在这里，我们对 740 头猪进行了测序，结合我们之前发布的 149 条重测序数据，检索了 207 个重测序数据集，形成了一个由全球分布的野猪、原住民和具有谱系结构的精选猪组成的面板，总计来自 43 个品种的 1096 个基因组。结合其单倍体信息读取和谱系结构，我们准确地构建了一组 1874 个单倍体基因组，其中包含 41,964,356 个遗传变异。我们通过识别五个新的肌内脂肪含量基因座，进一步证明了其在 GWAS 中的宝贵应用，并通过将估计育种值的准确性提高了 36.7%，在基因组选择中进一步证明了其有价值的应用。在进化选择中，我们检测了长期平衡选择下的MUC13基因，以及猪身高正选择下的NPR3基因。我们的研究为稳健的选择特征检测提供了丰富的基因组变异，并为破译猪的复杂性状提供了准确的单倍型。