PTMOverlay

Capstone PTM alignment This pipeline has been successfully tested on Linux, Windows and some Mac systems. It is known to have issues on ARM-based M4 systems.

UV Environment (Recommended)

To set up the workflow, first clone the git repository:

git clone https://github.com/evergreen700/PTMOverlay
cd PTMOverlay

If UV is not installed:

pip install uv

Create the UV environment:

uv sync

Run the workflow:

uv run snakemake --cores all

The final output files will be in the folder align_reports

Native

There are extensive packages to install, so a conda environment is recommended. Instructions for creating a new conda environment can be found here: https://docs.conda.io/projects/conda/en/latest/user-guide/tasks/manage-environments.html

Install dependencies using pip:

pip install matplotlib pandas pyteomics biopython lxml snakemake ncbi-taxonomist svgutils six
brew install pdf2svg (Mac)
sudo apt-get install pdf2svg (Linux)

Clone the git repository:

git clone https://github.com/evergreen700/PTMOverlay

We have not extensively tested all the packages this tool requires. These are the versions we have used:

• pyteomics: 4.7.5
• biopython: 1.85
• lxml: 5.3.1
• snakemake: 9.1.1
• ncbi-taxonomist: 1.2.1

MUSCLE Executable

In the executables directory of the repository, check to see if a MUSCLE executable that works with your operating system is present. If not, go to this website and download the correct executable and place it in in executables directory.

https://drive5.com/muscle/downloads_v3.htm

Then edit the runMUSCLE.py file accordingly.

If on Windows:

• Edit line 13 to point to the correct executable

If on Linux:

• Edit lines 16, 17 or 19, 20 depending on your machine

If on Mac:

• Edit line 23 or 25 depending on your machine

Execution

To run the workflow, place proteomes and kegg annotation files in the folder designated as proteome_dir in the config file. Place .pepXML files in folders sorted by ptm type within the folder designated as pepXML_dir in the config file. Below is an example of the file structure:

PTMOverlay
+ proteome
| + GCA_002847685.2.faa
| + GCA_002847685.2.kegg.txt
| + ...
+ mass_spec
| + Phospho
| | + BioD_urine_UMB0005_01_12Apr24_Arwen_WBEH-23-02-03.pepXML
| | + ... (other phospho pepXMLs)
| + ... (other PTM types)
+ index_umb_taxa_gca.tsv
+ README.md
+ snakefile
+ ...

index_umb_taxa_gca.tsv is a tab-separated file that is used to match between mass spec strain IDs (UMB####), species name, and proteome assembly (GCA). If you are using your own mass spec and proteome files, make sure that the names are in this tsv.

To run the workflow on your operating system:

cd PTMOverlay
snakemake --cores all

Example run

The kegg annotation files are included as an example. The proteomes and the mass spec search results are downloaded automatically from the MassIVE ftp server (see ftp_credentials.yaml) when running snakemake: download_sequence_ftp fetches one .fasta per assembly into proteome/, and download_mass_spec_ftp fetches one directory per search into mass_spec/, unpacking the .pepXML files out of the downloaded zip archives. Until the data is public, edit the ftp_credentials.yaml file with your username and password to access the data before running the workflow. Budget ~10 GB of free disk for the unpacked mass spec data. We recommend keeping the config file the way it is for the first run. If there are other orthologs or pathways you want to look at on the 31 species, rerunning with modified parameters will run faster if intermediates are already generated.