title	emoji	colorFrom	colorTo	sdk	sdk_version	app_file	pinned	license
PFLink	📊	green	blue	gradio	6.5.1	app.py	false	mit

Click here to use the website.

Files needed to run PFLink

• HDX Workbench
  • The default file export.
• HDExaminer
  • Two files are needed: "pool peptide results" (CSV) and "pool spectra export" (folder).
• DynamX
  • The default file export.
• BioPharma Finder
  • HDXData.csv
• Custom
  • The default custom file is found in the test_data directory of this repository.

---

Test data files

Example datasets are available in the test_data directory in this repository.

All ecDHFR data are from Pigeon-Feather.

• SpecExport.zip contains the isotopic mass envelope data for the ecDHFR dataset. This was generated from HDExaminer (pool spectra export).
• ecDHFR_tutorial.csv contains the isotopic uptake data for the ecDHFR dataset. This was also generated from HDExaminer (pool peptide results).
• flag-DHFR.txt contains all the protein and experimental metadata needed to generate the HXMS file.
• rangeslist.csv contains the peptides that should be kept when using PFLink.
• custom_empty.csv is an empty PFLink file that a user can fill in if they are not using a supported program.
• custom.csv is a filled example of the custom format with fake data values, as an example.

---

Installing and running PFLink locally

Installing via Conda

git clone https://huggingface.co/spaces/glasgow-lab/PFLink
cd PFLink
conda create -n PFLink python=3.11
conda activate PFLink
pip install gradio
pip install git+https://github.com/glasgowlab/PIGEON-FEATHER.git
pip install numba
pip install pyopenms
pip install hdxrate
pip install rich-argparse

Running the website locally via Conda

conda activate PFLink
cd PFLink # if not already in the directory
python3 app.py

Running the website locally via Docker

If you prefer running PFLink from the Docker rather than the local Conda installation, skip the instructions above and do this instead:

docker run -it -p 7860:7860 --platform=linux/amd64

registry.hf.space/glasgow-lab-pflink:latest python app.py

Running the scripts locally via Conda

conda activate PFLink
cd PFLink # if not already in the directory
python3 pflink_isotopic_uptake.py # for non-envelope data processing
python3 pflink_envelope.py # for envelope data processing
python3 pflink_merge.py # for merging different .HXMS files

---

Using the isotopic_uptake data processing script

This script is meant to convert DynamX, HDX Workbench, HDExaminer, BioPharma, or our custom export format into an .HXMS file. This is intended for centroid data only.

Arguments:

--help
Show this help message and exit.

--input_csv_path INPUT_CSV_PATH
Path to the file containing the raw input CSV/data file.

--output_hxms_path OUTPUT_HXMS_PATH
Path to the output directory where the final .HXMS files will be saved.

--flags_file_path FLAGS_FILE_PATH
Path to a flags or configuration file to override default settings.

--peptide_list PEPTIDE_LIST
Path to a CSV/TXT file containing a list of peptides to filter or process.

--saturation SATURATION
D2O saturation percentage (0.0 to 1.0) to record in the file header.

--ph PH
The measured pH (pH Read) value of the exchange buffer.

--temperature TEMPERATURE
The temperature of the exchange reaction in Kelvin (K).

--protein_name PROTEIN_NAME
The name of the protein.

--protein_state PROTEIN_STATE
The experimental state of the protein (e.g., 'Apo', 'Bound', 'Mutant').

--protein_sequence PROTEIN_SEQUENCE
The full amino acid sequence of the protein.

--include_exclude {include,exclude}
Determines if the peptide list file should be used to include or exclude peptides from processing.

--file_type {DynamX,HDXworkbench,HDExaminer,BioPharma,Custom}
Input CSV format type

**`input_csv_path`** and **`output_hxms_path`** are strictly required. However, you must either provide a complete flags file or use all options. The only two optional parameters are `include_exclude` and `peptide_list`.

Command usage:

python3 pflink_isotopic_uptake.py --input_csv_path {path} --output_hxms_path {path} --flags_file_path {path}

or

python3 pflink_isotopic_uptake.py --input_csv_path {path} --output_hxms_path {path} --saturation {saturation} --ph {ph} --protein_name {name} --protein_state {state} --protein_sequence {sequence} --file_type {file_type}

Examples of the flags file and peptide list can be found in the test_data folder.

---

Using the isotopic envelope data processing script

This script is meant to convert the HDExaminer export format into an .HXMS file. This is intended for HDExaminer envelope data only.

Arguments:

--help
Show this help message and exit.

--input_csv_path INPUT_CSV_PATH
Path to the file containing the raw input CSV/data file.

--output_hxms_path OUTPUT_HXMS_PATH
Path to the output directory where the final .HXMS files will be saved.

--flags_file_path FLAGS_FILE_PATH
Path to a flags or configuration file to override default settings.

--peptide_list PEPTIDE_LIST
Path to a CSV/TXT file containing a list of peptides to filter or process.

--raw_spectra_path RAW_SPECTRA_PATH
Path of the HDExaminer spectra folder.

--saturation SATURATION
D2O saturation percentage (0.0 to 1.0) to record in the file header.

--ph PH
The measured pH (pH Read) value of the exchange buffer.

--temperature TEMPERATURE
The temperature of the exchange reaction in Kelvin (K).

--protein_name PROTEIN_NAME
The name of the protein.

--protein_state PROTEIN_STATE
The experimental state of the protein (e.g., 'Apo', 'Bound', 'Mutant').

--protein_sequence PROTEIN_SEQUENCE
The full amino acid sequence of the protein.

--include_exclude {include,exclude}
Determines if the peptide list file should be used to include or exclude peptides from processing.

--save_match
Whether to save the match data. Default is False.

--save_fine_match
Whether to save the fine match data (uncentroided, fine structure). Default is False.

--file_type {HDExaminer}
Input CSV format type

input_csv_path, output_hxms_path, and raw_spectra_path are strictly required. However, you must either provide a complete flags file or use all options. The only two optional parameters are include_exclude and peptide_list.

Command usage:

python3 pflink_envelope.py --input_csv_path {path} --output_hxms_path {path} --raw_spectra_path {path} --flags_file_path {path}

or

python3 pflink_isotopic_uptake.py --input_csv_path {path} --output_hxms_path {path} --raw_spectra_path {path} --saturation {saturation} --ph {ph} --protein_name {name} --protein_state {state} --protein_sequence {sequence} --file_type {file_type}

Examples of the flag file and peptide list can be found in the test_data folder.

---

Using the HXMS merge script

This script is for combining replicates -- different experimental .HXMS runs of the same protein or state.

Arguments:

--help
Show this help message and exit.

--input_hxms_path1 INPUT_HXMS_PATH1
Path to the first file containing the .HXMS data.

--input_hxms_path2 INPUT_HXMS_PATH2
Path to the second file containing the .HXMS data.

--output_hxms_path OUTPUT_HXMS_PATH
Path to the output directory where the final .HXMS files will be saved.

All options are required to run the script.

Command usage:

python3 pflink_merge.py --input_hxms_path1 {path} --input_hxms_path2 {path} --output_hxms_path {path}

---

Using the website version on HuggingFace

Example datasets

• Use the "Isotopic uptake data load example" to load the ecDHFR dataset for isotopic uptake only.
• Use the "Envelope data load example" to load the ecDHFR dataset for both isotopic uptake and envelopes.

Once selected, all fields will be autofilled with the correct experimental and protein information.

Click the "Process Data" button at the bottom to generate the HXMS output file.