Manuscript Preprint: PubMed: 41000811
Data Repository: Silverman-Lab/mutt
This repository contains the R analysis code, helper functions, and R Markdown workflows used in the manuscript "Modeling Scale Uncertainty Improves Robustness of Microbiome Analysis."
It supports the full analytical pipeline — from data preprocessing and model comparison to statistical uncertainty estimation — used to evaluate microbiome datasets under various compositional and technical noise assumptions. At this time (2025), this represents the largest analysis of scale methods (35 datasets) and the largest benchmark across differential abundance methods.
| Path | Description |
|---|---|
| ALDEx3/ | Scripts and resources for ALDEx3 differential abundance analyses. |
| functions.R | Core analysis functions shared across scripts. |
| helperfunctions.R | Utility and plotting functions for reproducibility and report generation. |
| microbial_load_predictor/ | Main microbial load prediction workflow (primary implementation). |
| microbial_load_predictor1/ | Alternate version or experimental branch of the microbial load predictor pipeline. |
| mlpmuttanalysis.Rmd | Primary Analysis Script for Model Evaluation -- Machine learning and uncertainty model performance analyses (publication-ready version). |
| studycharacteristics.R | Script summarizing dataset characteristics and metadata harmonization. |
| wirbel/ | Supporting files for Wirbel et al. datasets and reproducibility checks. |
| autism.Rmd | Autism Fecal Gut Microbiome Re-analysis. |
Refer to the top of each .Rmd file for full session information and reproducibility environment.
Clone this repository:
git clone https://github.com/Silverman-Lab/mutt.gitIf you use this code or data (and more specifically the mutt repository), please cite:
Konnaris, MA et al.
Modeling Scale Uncertainty Improves Microbiome Analysis.
Preprint: PubMed: 41000811