Unsupervised Machine Learning Techniques to Categorize Genomic Islands

unsupervised Machine Learning (ML) techniques are powerful tools in identifying similarity patterns and can be utilized to categorize the data into related groups. This study showcases the applications of unsupervised methods, namely hierarchical clustering to carefully determine related groups of newly identified genomic islands. Genomic islands (GIs) are mobile genetic elements integrated into bacterial chromosomes. GIs can impact evolution of bacteria for example by carrying virulence or metabolic genes [1]. Precisely identifying GIs calls for a sophisticated process which we have recently implemented as an already published tool called TIGER which stands for Target / Integrative Genetic Element Retriever (TIGER). TIGER identifies mobile DNAs in each genome and identifies genomic islands with high accuracy [2]. We have employed TIGER to identify approximately 130,000 GIs in E. coli bacteria. To identify the similarities among the E. coli GIs, the hierarchical clustering algorithms and our heuristics have successfully categorized the data into relevant groups. We have been able to identify related groups of GIs as well as singleton GIs. Our results provide a promising method for categorizing large DNA segments that can be compared using a similarity measure and be categorized into more precise clusters for further analysis.