Artificial Intelligence and Machine Learning for Prediction and Analysis of Genomic Islands

Abstract

The Genomic Islands (GIs) are the groups of gene clusters in bacterial genomes acquired by lateral gene rransfer (LGT). These Genomic Islands are significantly linked with bacterial pathogenicity, adaptations, and evolution. The ancestries of GIs and their association with virulence or pathogenicity factors in bacteriophages and bacteria could provide detailed genetic diversity to identify nucleotide sequences. Increasing evidence suggests that LGT is the prime cause of the transfer of virulence genes through transduction, transformation, and conjugation. However, most of the GIs are the main origin of novel genes for some bacteria. Therefore, the prediction of Genomic Islands and their analysis have gained attention in bacterial genomic sequence research. Although, recently, several bioinformatics tools have been developed for detecting these GIs. Providing researchers with diverse options for effectively identifying these GIs in a bacterial genome. However, rarely any one of them is effectively identifying precisely the complete function of GIs in the bacterial genome. Therefore, advanced algorithms of Artificial Intelligence and Machine Learning approaches are used to process large GIs.