Where biology
meets algorithms

Microbial Clinical Atlas (MCA) is a curated catalogue of Taxon Passports that summarizes clinically relevant microbial features in a standardized, microbiome-ready format. MCA is designed to support both microbiologists and newcomers to microbiome analysis by providing reproducible, versioned annotations that remain stable across time, projects, and analytical pipelines. What MCA offers is a set of structured, comparable Taxon Passports that enforce the same terminology and field definitions across organisms, so interpretation is not dependent on ad hoc phrasing or analyst experience. By pairing controlled vocabularies with versioned records, MCA supports transparent comparison between taxa and consistent reporting across studies, while remaining approachable for users building their first microbiome workflows.

This project aims to integrate and compare deconvolution-derived gene expression programs across heterogeneous cohorts by constructing correlation-based metagene networks and detecting shared, biologically coherent communities. Through a stepwise workflow that links NMF/cNMF outputs to network topology, community annotation, and usage-based activity summaries, sotk2 supports reproducible cross-cohort module interpretation and prioritization.

This document defines a standardized data management plan (DMP) to promote sustainable, transparent, and reproducible organization of research data across projects and users. By establishing consistent practices for data storage, directory structure, documentation, and lifecycle management, the DMP reduces storage redundancy, improves disk usage tracking, and minimizes time spent locating or duplicating data. Through clear separation of raw data, derived results, references, and active project files, and by discouraging common organizational pitfalls, the DMP supports automation-ready storage accounting, reliable backup and archiving workflows, and long-term data stewardship across heterogeneous research activities.

This project aims to systematically evaluate and optimize the experimental variables that determine successful and reproducible engraftment of human gut microbiota in germ-free mice. Through quantitative assessment of microbial and host microbiome profiles, we seek to establish evidence-based best practices for generating and validating human microbiota-associated models.

This project aims to enable robust comparison of taxonomic composition across long-read metagenomic samples by converting taxonomic classification outputs into a genome-aware, base-level abundance metric. By normalizing species-assigned bases by reference genome length, BASEN mitigates biases introduced by sequencing depth, database composition, and taxonomic rank, which commonly confound raw read counts and proportional summaries. Through a structured workflow that restricts classifications to species level, aggregates base-level evidence, and computes normalized relative abundance, BASEN supports reproducible, rank-agnostic interpretation of taxonomic profiles in Nanopore-based metagenomic analyses.