Abstract Number: OC 17.1
Meeting: ISTH 2021 Congress
Theme: Arterial Thromboembolism » Cardiovascular Risk Factors
Background: Coronary artery disease (CAD) genome-wide association studies (GWAS) typically focus on single nucleotide variants (SNVs), but many potentially associated SNVs fail to reach the GWAS significance. Rather than focusing on a few SNVs strongly associated with CAD, by considering multiple SNVs in a gene and multiple genes in a pathway, our ability to identify novel genes and causal pathways can improve.
Aims: To identify genes and pathways associated with CAD using results from non-synonymous autosomal genetic variants in the CARDIoGRAM Exome (CGEX) GWAS and further compare in an independent multi-ancestry Pan-UK biobank (PUBB) GWAS using gene-based association (GBA) and pathway-based analyses.
Methods: We used versatile gene-based association study (VEGAS2) and Multi-marker analysis of genomic annotation (MAGMA) to perform GBA on CGEX (n = 120,575) and PUBB (n = 442,574) summary statistics. We included only exonic SNVs and excluded regulatory regions. VEGAS2 and MAGMA ranked genes (n = 15,296) and pathways based on aggregated SNV test statistics. We used Bonferroni corrected gene and pathway significance threshold at 3.0×10-6 and 1.0×10-5, respectively. We also report the top one percent of ranked genes and pathways.
Results: 
Venn Diagram of top enriched VEGAS2 and MAGMA genes in CARDIoGRAM and Pan UK Biobank
| Gene pathway name | Rank (Total = 5528) |
Genes in pathway | Empirical P-value | Rank (Total = 5741) |
Genes in pathway | Empirical P-value |
| CARDIoGRAM | Pan UK Biobank | |||||
| Lipid pathway | ||||||
| GO:0055088_lipid_homeostasis | 1 | 16 | 1.0×10-6* | 12 | 18 | 8.0×10-6* |
| REACTOME_LIPID_DIGESTION_MOBILIZATION_AND_TRANSPORT | 7 | 15 | 2.0×10-6* | 4 | 17 | 2.0×10-6* |
| Coagulation | ||||||
| PANTHER_BIOLOGICAL_PROCESS_Blood_clotting | 8 | 14 | 8.0×10-6* | 38 | 29 | 1.9×10-4 |
| Inflammation | ||||||
| PID_AMB2_NEUTROPHILS_PATHWAY | 26 | 10 | 9.2×10-5 | 42 | 10 | 9.2×10-5 |
| Wound healing | ||||||
| GO:0042060_wound_healing | 47 | 36 | 4.2×10-4 | 882 | 51 | 7.0×10-2 |
| Neuronal aging | ||||||
| GO:0043523_regulation_of_neuron_apoptosis | 42 | 13 | 2.2×10-4 | 123 | 15 | 1.9×10-3 |
Top one percent VEGAS2 lipid and non-lipid based enriched pathways in CARDIoGRAM and Pan UK Biobank
We identified 17 top enriched genes (Figure) with four genes (PCSK9, FAM177, LPL, ARGEF26) reaching statistical significance (p ≤ 3.0×10-6) using both GBA approaches in two GWAS studies. In addition, our analyses identified 10 genes (DUSP13, KCNJ11, CD300LF/RAB37, SLCO1B1, LRRFIP1, QSER1, UBR2, MOB3C, MST1R, and ABCC8) with previously unreported associations with CAD, although no single SNV associations within the 10 genes reached genome-wide significance. Among the top one percent pathways, we detected pathways regulating lipid homeostasis, coagulation, inflammation, wound healing and neuronal aging (Table).
Conclusions: The VEGAS2 and MAGMA GBA lead to discovering previously unreported genes and mapping pathways associated with CAD. Through GBA we further identified multiple genes carrying sub GWAS threshold SNVs and pathways associated with CAD.
To cite this abstract in AMA style:
Hariharan P, Dupuis J. Mapping Gene and Gene Pathways Associated with Coronary Artery Disease: A CARDIoGRAM Exome and Multi-ancestry UK Biobank Analysis [abstract]. Res Pract Thromb Haemost. 2021; 5 (Suppl 2). https://abstracts.isth.org/abstract/mapping-gene-and-gene-pathways-associated-with-coronary-artery-disease-a-cardiogram-exome-and-multi-ancestry-uk-biobank-analysis/. Accessed December 11, 2023.« Back to ISTH 2021 Congress
ISTH Congress Abstracts - https://abstracts.isth.org/abstract/mapping-gene-and-gene-pathways-associated-with-coronary-artery-disease-a-cardiogram-exome-and-multi-ancestry-uk-biobank-analysis/