SNP-SNP Interaction Using Gauss Chaotic Map Particle Swarm Optimization to Detect Susceptibility to Breast Cancer

SNP-SNP interaction studies focus on the investigation of associated variations in disease susceptibility. In recent years, increased attention has been focused on the analysis of huge combinations of SNPs in association studies. The large number of calculations required, however, often limits the application of statistical methods. In this study, we applied an evolutionary algorithm to facilitate statistical methods in the analysis of associated variations for disease susceptibility. The Gauss particle swarm optimization (Gauss PSO) algorithm was used to detect and identify the best protective association (i.e., combinations of SNPs with genotypes) with breast cancer. In the simulated data set, we systematically evaluated the combination effects of 26 SNPs from eight epigenetic modifier-related genes for breast cancer. The 2- to 6-order SNP combinations were identified for protective association with the risk of breast cancer (odds ratio <; 1.0; p-value <; 0.05). Experimental results showed that genes, EGF (rs2237054), IGF2 (rs680), IGFBP3 (rs2471551, rs2132572), IL10 (rs1554286) and VEGF (rs3025039) were statistically significant, and played an important role in the interactive effects in breast cancer. We compared the methods of PSO and Gauss PSO to identify SNP-SNP interaction. Analysis results supported that the Gauss PSO was able to identify higher difference values than the PSO; the difference in the obtained values was used to distinguish between the breast cancer group and non-cancer group. The results revealed that the Gauss PSO was a robust and precise algorithm for identifying the best protective SNP combinations.