Benchmarking of Statistical Models and Machine Learning to Predict Taxa Contributing to Shannon Diversity in Organic Gut Microbiomes

The development of cheaper sequencing of genomic data has led to an abundance of that that can be utilized, however in that abundance of data its quality can often times be questionable. In this study, I utilized machine learning methods to find significant taxa in OTU tables based on models predicting Shannon diversity. A variety of methods were benchmarked, the most effective being the random forest model which is typically implemented in genomics.