Previous reports including genome-wide association studies (GWASs) have described associations of serum lipids with genomic variations. In the present study, we examined the association of ∼2.5 million single-nucleotide polymorphisms (SNPs) from 3041 Japanese healthy volunteers obtained from the Japan Pharmacogenomics Data Science Consortium (JPDSC) database with serum lipids. We confirmed the previously reported associations of 14 SNPs in 5 regions for low-density lipoprotein (LDL) cholesterol, 23 SNPs in 12 regions for high-density lipoprotein (HDL) cholesterol, 16 SNPs in 6 regions for triglyceride and 5 SNPs in 1 region for phospholipid. Furthermore, we identified 16 possible novel candidate genes associated with LDL cholesterol, HDL cholesterol or triglycerides, where SNPs had P-values of <1 × 10(-5). Further replication analyses of these genes with Korean data revealed significant associations of SNPs located within the PCSK7 gene and triglyceride (Pmeta=7.98 × 10(-9) and 1.91 × 10(-8) for rs508487 and rs236911, respectively). These associations remained significant even by the conditional analysis adjusting for three neighboring variations associated with triglyceride. Our present data suggest that PCSK7 as well as PCSK9 may be associated with lipids, especially triglyceride, and may serve as a candidate for a new drug target to treat lipid abnormality syndromes.