M. T. Ruffin

A framework for effective application of machine learning to microbiome-based classification problems

This paper evaluates ML models for predicting colonic neoplasias using 16S rRNA data, finding that L2-regularized logistic regression offers a balance of performance, speed, and interpretability.

Fecal short-chain fatty acids are not predictive of colonic tumor status and cannot be predicted based on bacterial community structure

This study finds that fecal SCFA concentrations are not associated with colonic tumor status and cannot be predicted from bacterial community structure.

The Diagnostic Potential & Interactive Dynamics of the Colorectal Cancer Virome

The paper examines the colorectal cancer virome and identifies differences in virome composition between healthy and cancerous samples, suggesting a potential role for bacteriophages in cancer progression.

Spatial variation of the native colon microbiota in healthy adults

The paper profiles the microbiota of the healthy colon to identify spatial variations between proximal and distal regions using 16S rRNA gene sequencing and classification models.

Normalization of the microbiota in patients after treatment for colonic lesions

This study examines microbiota changes in patients after treatment for colonic lesions, finding significant normalization in carcinomas but not adenomas or advanced adenomas.