Authors: Martha Powell, Future Science Group
Researchers from Washington State University (WA, USA) have developed a novel method for identifying antibiotic resistance genes in bacteria using game theory and machine learning.
In recent years genome sequencing has been used to identify antibiotic resistance genes by differentiating unknown sequences with known resistance sequences in existing online repositories; however, this method doesn’t hold up for new or unusual genes that have low sequence identity with known sequences. This study, published in Scientific Reports, used game theory, which models strategic interactions between game players, to identify antibiotic resistance genes.
The team employed game theory to choose which protein characteristics to use in the machine learning model. The team assessed interactions of several features of genetic material including structure, physiochemical, evolutionary and composition properties of protein sequences.