Efficient genetic improvement of complex traits in pigs requires a deep understanding of their genetic architecture and a comprehensive catalog of molecular markers. Here, we conducted a genome-wide association study (GWAS) and conditional analysis in 4,144 Yorkshire pigs, using 10.4 million imputed sequence variants and phenotype data on average backfat thickness at 100 kg (BFT_100). We identified 11 independently significant variants and prioritized 17 candidate genes. By integrating the Pig Genotype-Tissue Expression (PigGTEx) resources, we further performed a transcriptome-wide association study (TWAS) and colocalization analyses, revealing 27 TWAS-significant genes and 13 colocalized genes. These integrative strategies enhanced the functional signals capturing capability based on conventional GWAS, providing a deeper understanding of BFT_100 biology. Notably, NFIA, FCN1, and DBH emerged as novel candidate genes that are not previously implicated for similar traits in either the Animal QTLdb or PigBiobank. Our findings provide a reference for dissecting the genetic architecture of porcine BFT_100 and offer potential markers for precision breeding.