Coronary Artery Disease (CAD) is the leading morbid condition worldwide. It is the major health challenge for South Asians. The disease burden is even higher in Pakistan. Being a polygenic disorder, CAD pathogenesis involves multiple genes. Population based genetic variations in these genes, may influence the risk of CAD. This study aimed to assess the association of environmental/genetic risk factors with angiographically assessed/clinically determined CAD in Pakistani population. Genome wide association studies (GWAS) have implicated about 46 CAD loci associated with many variants but most of them lie in non-coding regions. Public databases have emerged to define the function of these variants. Assuming that some of implicated variants are associated with disease risk by affecting the gene regulation, we also determined the regulatory role of these single nucleotide polymorphism (SNPs) residing in the non-coding regions. A total of 695 subjects (22.3% female, mean age= 54 ± 11 years) were included. CAD stenosis/extent was assessed by angiography. The subjects were categorized as having severe CAD (≥70% stenosis in ≥1 vessel), moderate CAD (30-69% stenosis in at least one vessel) and no CAD (<30% stenosis). For genetic risk screening, we selected 47 genetic variants associated with 43 genetic loci. The subjects were also evaluated for APOE gene polymorphism. Genotypes of 47 variants were performed using the Sequenom iPLEX assay and APOE polymorphisms (E2/E3/E4) were determined using TaqMan assays. The association of genetic variants with coronary stenosis was determined by chisquare and additive genetic model. We used Regulome database (DB) to identify the regulatory variants among 1,121 CAD associated SNPs and their tagged SNPs. Functional annotation of significant SNPs was determined using RegulomeDB and SNAP web portal databases. Among environmental risk factors, Low density lipoprotein cholesterol (LDLC) and hypertension appeared as significant (p<0.034 and p<0.011 respectively) nongenetic risk factors in our population. We had five significant SNPs after dominant model analysis; (PLG/rs4252120; p=0.003, KIAA1462/rs2505083; p=0.006, LPA/rs2048327; p=0.04, SORT1/rs602633; p=0.02 & UBE2Z/rs46522; p=0.02). Of these top 5 variants, two of them; PLG/rs4252120 (p=0.003) and KIAA1462/rs2505083 (p=0.006) showed significant association with CAD in our sample even after correcting for multiple testing using false discovery rate (q<0.05). The Odds ratio (OR) in patients Vs. controls for two significant SNPs were; [rs4252120 (OR=1.83; p=0.003, FDR=0.02) & [rs2505083 xv (OR=1.65, p=0.006, FDR=0.03)]. For APOE gene polymorphism 672 subjects were successfully genotyped. The frequency of APOE*4 carriers (3/4 and 4/4 genotypes) was significantly higher in severe stenosis group (≥70%) as compared to control group (<30%) (22.8% Vs. 13.01%; p=0.01). In multiple regression, the odds ratio for APOE*4 carriers to develop >70% stenosis was 2.16 (95% CI: 1.29-3.79; p<0.005). Out of 1121 GWAS significant and tagged SNPs, 790 returned a numeric RegulomeDB score of 1-6, while remaining variants had no data. Only 90 were strongly predicted as regulatory SNPs with a score <3 and 8 of them were GWAS significant; LIPA/rs2246833(RegulomeDB score=1b), ZC3HC1/rs11556924, CYPA1/CNNM2/NT5C2/rs124113409, APOE-APOC1/rs2075650, and UBE2Z/rs46522 (RegulomeDB score=1f), ZNF259-APOA5-APOA1/rs964184, UBE2Z/rs46522, SMG6/rs2281727, and COL4A1-COL4A2/rs4773144 (RegulomeDB score= 2b). In conclusion, LDL-C and hypertension were found as significant risk factors. We successfully replicated 2 previously reported genome-wide significant SNPs among Europeans in our Pakistani sample. PLG/rs4252120 & KIAA1462/rs2505083 are significant risk factors for CAD in Pakistanis. Our study also determined that the presence of APOE*4 allele is a risk allele to develop severe coronary stenosis (>70%) among Pakistanis. This study fosters that some of non-coding genetic variants are true signals and regulate the gene expression at transcriptional level. Our study indicates that RegulomeDB is a useful database to examine the large number of genetic variants and to differentiate between true or tagged SNPs after defining the functional role of variants, residing in GWAS-implicated loci.