Using neural networks to mine text and predict metabolic traits for thousands of microbes

Microbes can metabolize more chemical compounds than any other group of organisms. As a result, their metabolism is interest to investigators across biology. Despite the interest, information on specific microbes hard access. Information buried in text books and journals, have no easy way extract it out. Here we investigate if neural networks out this predict metabolic traits. For proof concept, predicted two traits: whether carry one type (fermentation) or produce metabolite (acetate). We collected written descriptions 7,021 species bacteria archaea from Bergey’s Manual . read manually identified (labeled) which were fermentative produced acetate. then trained these labels. In total, 2,364 as fermentative, 1,009 also producing Neural could with 97.3% accuracy. Accuracy was even higher (98.6%) when predicting Phylogenetic trees traits confirmed that predictions accurate. Our approach efficiently accurately. It paves for putting into databases, providing access investigators.