Published October 15, 2026 | Version v4

Source Data and MAGs for FMT-HIV Butyrate Analysis

  • 1. The Affiliated Hospital of North Sichuan Medical College, Nanchong, China

Description

Version v4: Data and code supporting the HIV-FMT butyrate terminal-marker analysis

This release provides the data, analysis scripts, and supplementary results supporting a secondary genome-resolved analysis of 84 shotgun metagenomic samples from 28 participants in an HIV fecal microbiota transplantation (FMT) study. The analysis uses 279 non-redundant metagenome-assembled genomes (MAGs) and compares two terminal-marker definitions: But-supported and atoAD-expanded.

All 84 samples were quantified competitively against the same unified 279-MAG reference using Salmon. TPM-derived MAG abundance proxies were used for the primary analyses, and NumReads weights were used for the LMM abundance-scale sensitivity analysis.

Files included

1. Unified reference
all_279_MAGs_unique_contigs.fasta.gz
Gzip-compressed FASTA containing the unified 279-MAG reference used for competitive read quantification. The SHA-256 checksum of the decompressed FASTA is:
6193b7eabeb0cbd5c757aaff346da54da74b06975411b0800eefa920a403721c

2. Final abundance matrices
MASTER_MAGs_Abundance_Matrix_REQUANTIFIED.csv
MASTER_MAGs_Abundance_Matrix_NUMREADS.csv
Each matrix contains 279 MAGs and 84 samples. These files provide the TPM-derived abundance proxies and NumReads weights, respectively.

3. Marker scores and candidate assignments
MAG_butyrate_capacity_strict_expanded.csv
MAG-level terminal-marker scores under the But-supported and atoAD-expanded definitions.

but_candidate_assignment.csv
Sequence/phylogeny-supported classification results for 452 candidate proteins.

4. Analysis scripts
run_all.R
run_all_NUMREADS.R
build_matrix.py
build_sample_QC.py
but_classification_pipeline.sh
but_classification.R

These scripts cover the TPM-based analyses, NumReads-weighted LMM sensitivity analysis, abundance-matrix construction, sample-level QC summary generation, and But-candidate classification. Input requirements and execution details are described in README_v4.txt.

5. Supplementary results
supplementary_v4.tar.gz

This archive contains:
- Supplementary Figure S1: Salmon mapping-rate distribution.
- Supplementary Table S1: Sample-level quantification QC.
- Supplementary Table S2: ENA run/sample crosswalk.
- Supplementary Table S3: MAG taxonomy.
- Supplementary Table S4: But-candidate assignment.
- Supplementary Data S5: TPM- and NumReads-weighted LMM fixed-effect tests and visit-specific between-group contrasts, together with the descriptive correlation between marker scores.
- Supplementary Data S6: W08 Top15 MAG lists and paired patient-level bootstrap results.
- Supplementary Data S7: Exploratory nadir CD4 analyses.
- Supplementary Data S8: MAG contribution turnover analyses.
- The MAG-level terminal-marker score table and an archive README.

6. Documentation and verification
README_v4.txt
sessionInfo_TPM.txt
sessionInfo_NUMREADS.txt
SCRIPT_SHA256SUMS.txt
SHA256SUMS.txt

The session information files document the R environments used for the analyses. SCRIPT_SHA256SUMS.txt provides checksums for the analysis scripts. SHA256SUMS.txt provides checksums for the distributed files, including the compressed reference FASTA.

Interpretation

The abundance values are DNA shotgun-based MAG abundance proxies. They do not represent RNA transcript abundance, absolute MAG copy number, measured butyrate production, metabolic flux, or clinical efficacy.

But-supported assignments indicate sequence/phylogeny-supported candidate butyryl-CoA:acetate CoA-transferase classification, rather than experimentally validated enzyme function. The atoAD-expanded definition is a marker-definition sensitivity analysis and should not be interpreted as butyrate-specific functional evidence.

Version information

Version v4 supersedes v1-v3 as the release supporting the associated manuscript. It consolidates the unified-reference matrices, analysis scripts, and supplementary results. The earlier abundance matrix generated using a single-patient P01 assembly as the Salmon reference, and results derived from that matrix, were excluded from the analyses reported in the manuscript.

Raw sequencing reads are available through the European Nucleotide Archive under BioProject PRJEB75958.

Files

but_candidate_assignment.csv

Files (163.8 MB)

Name Size
md5:f5efdb43b82b433bc4324f9f1b952385
163.0 MB Download
md5:e723ba3caa1f7b163c6bc992346d8802
1.4 kB Download
md5:06fac8a78787812cbdd08677a875b62d
1.3 kB Download
md5:b2f8fa2e7b987ba4b475ca0e2930b29c
36.4 kB Preview Download
md5:91461bd1f202fe40b1b907e0c15b0f8b
1.1 kB Download
md5:66e54ce8a780716a79cadf27a7cc063e
889 Bytes Download
md5:ae32b9030fa1703f5e60b77cf95fd7a7
9.9 kB Preview Download
md5:05f94317dd96e1037783d20444166eaa
291.9 kB Preview Download
md5:e49599777f03df984f0798f8e75b0e44
364.7 kB Preview Download
md5:142ad64bf615004915a7b7bda912c330
4.8 kB Preview Download
md5:c74f75f15ad8a06a792f683f54263b89
14.0 kB Download
md5:a1da30cfe2acd3f932257a3d502189fd
3.3 kB Download
md5:0017066c587980c68a442d76ea7dc77a
512 Bytes Preview Download
md5:d73e1d7884130ad76a9f4c0d0a8c849e
1.8 kB Preview Download
md5:e45a983cfbf6833341b7e732427b9890
1.7 kB Preview Download
md5:a66d3c50c8006bfd5ccf4058a23a3a37
1.5 kB Preview Download
md5:555e41d5e45aa4eac4b1bfc1a3e9fa34
27.1 kB Download