19. Data Mining 16S rRNA Gene Sequences from Antibiotic-Producing Bacteria

Data Mining 16S rRNA Gene Sequences from Antibiotic-Producing Bacteria

Renata Rigueira, James Clark, Rachel St. Clair, Timothy Stinson, Elan Barenholtz, Diane Baronas-Lowell
Florida Atlantic University

Soil bacteria are isolated and tested for killing of nine safe relatives to ESKAPE pathogens, as well as, resisting five common antibiotics. A neural network is being used to discover novel patterns in the 16S rRNA DNA sequence that may be responsible for antibiotic synthesis or resistance. The 16S rRNA gene sequences from bacterial isolates that kill Pseudomonas putida are used to train the network to find patterns that may be responsible for antibiotic synthesis. The 16S rRNA DNA sequences of bacterial isolates are also analyzed for penicillin resistance to look for a correlation between the two. Preliminary testing conducted with 218 bacterial isolates showed the neural network was able to predict, with an 80% accuracy rate, which bacteria would kill Pseudomonas putida based on their 16S rRNA DNA sequence. Although, the accuracy rate needs improvement, its hoped that more data may improve the accuracy rate of the network.