Sovereign 8s.is mirror of standardgalactic/mitch_manuscript
Find a file
Repository files (latest commit first)
Filename Latest commit message Latest commit date
2026-03-02 14:32:07 +01:00
scripts first commit 2026-03-02 14:29:31 +01:00
.gitignore first commit 2026-03-02 14:29:31 +01:00
README.md Fix formatting of author contribution notes in README.md 2026-03-02 14:32:07 +01:00

MitCH Manuscript: Multi-omics Analysis of Bacterial Vaginosis

Study

Bacterial vaginosis as a metabolic state: multi-omics integration reveals community-driven polyamine signatures independent of Gardnerella strain diversity

Hana Shabana*, Faruk Dube*, Emilia Lahtinen, Emelie Bergman Rothe, Lars Engstrand, Luisa W. Hugerth, Ina Schuppe-Koistinen#, Kenny A. Rodriguez-Wallberg#

*Equal contribution as first authors; #Equal contribution as senior authors

Description

Companion code and processed data for the MitCH multi-omics BV study. This repository contains the analysis scripts, processed intermediate objects, and supplementary materials for a cross-sectional study of 111 Swedish women (29 BV-positive, 82 controls) integrating shotgun metagenomics, targeted LC-MS metabolomics, and Gardnerella pangenomics.

Data Availability

Data type Location Status
Raw sequencing reads ENA accession PRJEB108308 Available upon publication
Processed objects and metadata This repository Available
Large supplementary files Zenodo (DOI to be added) Available upon publication
Raw metabolomics data Available from authors upon request -

Repository Structure

mitch_manuscript/
├── README.md
├── data/                           # Gardnerella pangenome gene presence/absence
├── metadata/
│   └── sample_metadata.rda         # Sample metadata (n=111)
├── objects/                        # Cached R objects for reproducibility
│   ├── multiomics_views.rda        # 4-view omics data (metabolomics+/-, pathways, taxonomy)
│   ├── group_labels.rda            # BV/Control group labels
│   ├── diablo_selected_features.rda # DIABLO feature selection results
│   └── genome_metabolome_results.rda # Genome-metabolome integration results
├── scripts/
│   ├── 00_data_processing_and_DIABLO.R  # Raw data processing and DIABLO integration
│   ├── 01_MOFA2_integration.R           # MOFA2 multi-omics factor analysis
│   └── 02_genome_metabolome.R           # Genome-metabolome integration + Figure 7
├── figures/                        # Manuscript figures
└── supplementary_tables/           # Tables S1-S11

Script Execution

Execution order

00_data_processing_and_DIABLO.R  (requires raw data, not included)
         |
         v
   +-----+-----+
   |             |
   v             v
01_MOFA2       02_genome_metabolome
(independent)  (independent)

Script descriptions

Script Purpose Raw data needed?
00_data_processing_and_DIABLO.R Processes raw sequencing and metabolomics data, performs differential abundance (ANCOM-BC2, limma), and fits DIABLO multi-omics integration Yes (see script header for data sources)
01_MOFA2_integration.R Unsupervised multi-omics factor analysis (MOFA2) with 4 data views No (uses cached objects)
02_genome_metabolome.R Gardnerella pangenome-metabolome integration (Mantel, Procrustes, db-RDA) and publication Figure 7 No (uses cached objects)

Reproducibility note: Scripts 01 and 02 are fully reproducible using the cached objects in objects/ and metadata/. Script 00 documents the complete upstream pipeline but requires raw data files available from ENA and from the authors.

R Dependencies

Bioconductor packages

MOFA2, mixOmics, ANCOMBC, DESeq2, phyloseq, limma, decontam, impute, sva, BiocParallel, ALDEx2, Maaslin2

CRAN packages

tidyverse, dplyr, tidyr, purrr, readr, tibble, magrittr, stringr, readxl, reshape2, data.table, ggplot2, ggrepel, ggtext, pheatmap, corrplot, RColorBrewer, cowplot, patchwork, gridExtra, scales, vegan, ade4, compositions, randomForest, caret, openxlsx, broom, matrixStats, MetaboAnalystR

Installation

# Bioconductor
if (!require("BiocManager")) install.packages("BiocManager")
BiocManager::install(c("MOFA2", "mixOmics", "ANCOMBC", "DESeq2", "phyloseq",
                        "limma", "decontam", "impute", "sva", "BiocParallel",
                        "ALDEx2", "Maaslin2"))

# CRAN
install.packages(c("tidyverse", "readxl", "reshape2", "data.table", "vegan",
                    "ade4", "compositions", "ggrepel", "ggtext", "pheatmap",
                    "corrplot", "RColorBrewer", "cowplot", "patchwork",
                    "gridExtra", "randomForest", "caret", "openxlsx",
                    "broom", "matrixStats", "MetaboAnalystR"))

Ethics

This study was approved by the Swedish Ethical Review Authority (reference: 2021-05794-01). All participants provided written informed consent.

Citation

Manuscript under review

License

MIT