This repo contains the analysis pipeline for this study.

Notebooks containing the current results of data analysis. These contain the figures and plots related to this study.

Contains scripts used in data analysis pipeline. Primaryly writen in Python3, most are borrowed from the Metaprotemics package to which I also contriubte.

Collection of useful bash 1-liners for moving data around, starting various anaoysis piplnes and performing basic operations (e.g. in-place text replacement).

Samples were generated by extracting proteome from mouse fecal samples. Microbiome metaproteome was mixed 1:1 heavy to light (N15:N14). Samples were analyzed via LC/MS/MS on an Thermo LTQ Orbitrap XL via 11 step MudPIT. MS1 and MS2 files were extracted via rawXtract and searched with blazmass against the ComPIL MGM database on the Scripps HPC garibaldi. Search results were filterd via DTASelect to a 0.01 FDR at the protein level. For the purposes of this repo, raw data files are missing, but they are to be placed in a top-level folder data to perform the analysis contained here.

N14 - RAG1 KO/Tcell Transfer Mice: These mice develop symptoms of inflammatory bowel disease (IBD) within weeks of Tcell transfer.

N15 - RAG1 KO Mice: These mice have no immune cells, but are otherwise healthy in labratory conditions.

11-Step MudPIT

Data extracted with rawXtract v1.9.9.2 using the data-dependant option into .ms1 and .ms2 files. Files were uploaded to the Scripps HPC Garibaldi for search.

Search queueing was handeled by the Metaprotemics package written by Su lab for both handling submission of search jobs on the HPC cluster and downstream analysis. Several scripts in the scripts/ directory are borrowed from this package.

Search results were filtered with DTASelect 2.0 using the following options:

• trypstat - use cleavage status when calculating probabilities
• modstat - Use separate models for modified peptides
• y 2 - Include only fully cleaved peptides
• DM 10 - Set highest delta mass at 10ppm
• pfp 0.01 - protein filtered to a 1% FDR
• p 2 - require a minimum of 2 peptides per protein identified