Usage

Install xlranker using pip or other Python package managers.

pip install xlranker

This will install a xlranker command which can be used to run the pipeline. You can also use the library if you are using a Jupyter Notebook. For notebook users, please see the notebook example.

Input Data

The input data for xlranker are:

Peptide Pairs

TSV file showing all of the identified Peptide Pairs in the dataset.

Omic Data

Omic data used by the machine learning model for prioritizing ambiguous pairs

Custom Sequence Mapping (Strongly Recommended, Optional)

By default, xlranker uses the human UNIPROT (accessed 5-30-2025) one sequence per gene to map peptide sequences to proteins. It is strongly recommended that you provide the same database used for mapping the proteomics data. You can provide either a FASTA file or a TSV table with mapping pre-computed Please read documentation for requirements.

The typical file structure for the input looks like

omic_data/
├── protein.tsv
└── rna.tsv
peptide_network.tsv

Running the Pipeline

Example Data

To test the pipeline or view the input data formatting, download the example data below

Download example.tar.gz

For most users, you would want to run the full pipeline. This can be achieved by running the following command:

xlranker start peptide_network.tsv omic_data/

This example assumes peptide_pairs.tsv is already prepared according to the instructions above and is in the current working directory.

The CLI contains multiple feature flags, such as only using the parsimony selection, saving more data, and custom filtering options. To view all of the options, please see CLI option documentation

Output

The output of the pipeline contains two files and a folder

model/

Folder containing data from the machine learning model.

reports/

Folder containing protein networks at different confidence levels.