pepMAP

pepMAP is a web-based application for visualizing peptide mappings onto protein sequences. It supports input from FragPipe, DIA-NN, MaxQuant, and Spectronaut reports, providing an interactive interface to explore peptide coverage and protein features.

Features

• Support for DIA-NN, FragPipe, MaxQuant, and Spectronaut Inputs: Upload peptide reports to visualize peptide mappings on proteins.

• Interactive Visualization: Explore peptide coverage with an intuitive interface powered by Plotly.

• Protein Feature Integration: Fetch and display protein domains, binding sites, modified residues, and variants from UniProt.

• Multi-User Sessions: Supports multiple users simultaneously with isolated sessions.

• Custom Features Upload: Accepts custom features either with the initial upload or in a later step.

• API Access: Provides endpoint for uploading data and plotting.

• Peptide Summary Tables: Generate summary tables of peptide coverage per sample, topology, or domain.

• Sample Selection + Regex Cleanup: Select a subset of samples and optionally clean sample names with a regex.

Installation

To run pepMAP locally, follow these steps:

1. Clone the Repository

git clone https://github.com/npinter/pepMAP.git
cd pepMAP

2. Create a Conda Environment

Create the Conda environment from the included environment.yaml:

conda env create -f environment.yaml
conda activate pepMAP

3. Run the Application

Activate the Conda environment and start the Flask application:

conda activate pepMAP
(pepMAP) python app.py

By default, the application runs on http://localhost:7007. Open this URL in your web browser to access pepMAP.

Usage

1. Upload Files: Upload your peptide report and FASTA file used in the search.

DIA-NN: report.tsv/.parquet
FragPipe: psm.tsv
MaxQuant: evidence.txt
Spectronaut: report.tsv

FASTA: db.fasta

2. Search Proteins: Enter a UniProt ID or gene symbol to visualize peptide mappings.

3. Custom Features Format: Use a tab-separated file with the columns uniprot, position, description, literature. The position field can be a single number or a range.

Example (custom_features.tsv):

uniprot position    description literature
P04626  1242-1254   IHC (Ventana 4B5)   https://doi.org/10.1111/j.1365-2559.2011.04034.x
P04626  557-603 ADC (Trastuzumab/Herceptin) https://doi.org/10.1038/nature01392
Q9NZQ7  284-290 IHC (Ventana SP263) https://doi.org/10.1038/s41379-019-0372-z

4. API Usage (Upload + Plotting): The API supports file uploads with an optional session_id, and plotting endpoints accept the same settings as the UI (proteotypic, charge states, q-value (EP) cutoff, sample name cleanup, summary table mode, and optional sample selection via selected_runs). The browser UI creates a session on upload, while external tools can supply their own. Session IDs must be 8-128 characters and match [A-Za-z0-9_-].

Endpoints

• POST /upload (multipart): report_file, fasta_file, organism (HUMAN/MOUSE/CUSTOM + custom_organism), optional custom_features_file + custom_features_label, optional session_id to reuse, and optional plot defaults: proteotypic_checkbox (true/false), charge_state_mode (all/unique), q_value_cutoff (0-0.05), sample_name_cleanup (none/split_underscore/custom), sample_name_custom_pattern (regex; capture group 1 used if present), summary_mode (per_sample/per_topology/per_domain).
• POST /plot_peptides and POST /plot_features (multipart): require search_input (gene symbols and/or UniProt IDs separated by spaces/commas/semicolons), accept optional search_labels (JSON array of custom names aligned with search_input), and support the same settings plus optional selected_runs (JSON array of sample names).

Example (external upload + plotting with curl):

# upload report + fasta and optional custom features, get or reuse a session_id
curl -X POST http://127.0.0.1:7007/upload \
  -F session_id=my_session_001 \
  -F report_file=@report.tsv \
  -F fasta_file=@db.fasta \
  -F organism=HUMAN \
  -F custom_features_file=@custom_features.tsv \
  -F custom_features_label="IHC Panel" \
  -F proteotypic_checkbox=true \
  -F charge_state_mode=unique \
  -F q_value_cutoff=0.01 \
  -F sample_name_cleanup=custom \
  -F sample_name_custom_pattern='^([^_]+)' \
  -F summary_mode=per_topology

# plot one or more genes/proteins and optionally set custom display names
curl -X POST http://127.0.0.1:7007/plot_peptides \
  -F session_id=my_session_001 \
  -F search_input='TP53 EGFR' \
  -F search_labels='["p53 custom","EGFR custom"]' \
  -F selected_runs='["Sample_A","Sample_B"]'

License

This project is licensed under the MIT License.

Note

• Ensure that you have an active internet connection, as the application fetches protein features from the UniProt API.
• The application stores uploaded data in storage/ and schedules daily deletion to manage server resources effectively.