Skip to content
This repository has been archived by the owner on Jul 1, 2022. It is now read-only.
/ fs-tool Public archive

Tool to calculate fraction of shared bound peptides between HLA proteins

License

Notifications You must be signed in to change notification settings

bjohnnyd/fs-tool

Repository files navigation

fs-tool

Release Build Status Software License

Command-line tool to calculate fraction of shared bound peptides between HLA alleles from NetMHCpan binding predictions. The tool currently reports fraction of shared peptides based on default motifs in the peptide but additional positions can be supplied by the user.

Current HLA allele ligand group assignments that is included with this tool was obtained from https://www.ebi.ac.uk/ on 2019-12-29. The kir ligand motifs can be updated using the tool.

Detailed instructions and descriptions are available in the Documentation.

Install

Binary

The simplest way to install is using the precompiled binaries provided below:

picture picture picture
TAR TAR TAR
ZIP ZIP ZIP

Using the command line you can obtain the binary using (Linux):

$ curl -sL https://github.com/bjohnnyd/fs-tool/releases/latest/download/x86_64-unknown-linux-gnu.tar.gz | tar xvz && chmod +x fs-tool && ./fs-tool -h

Build

To compile from source rustup is required and can be obtained HERE. After installing rustup download the release archive file and build:

$ git clone https://github.com/bjohnnyd/fs-tool.git && cd fs-tool && cargo build --release --bin fs-tool

The compiled binary can then be ran using:

$ ./target/release/fs-tool -h

All releases and associated binaries and archives are accessible under Releases.

Usage

Running fs-tool -h will list all possible arguments:

$ ./fs-tool -h

fstool 0.2.5
Calculates fraction of shared bound motifs between HLA alleles while incorporating KIR ligand and LILRB binding
information.

USAGE:
    fs-tool [FLAGS] [OPTIONS] --binding-predictions <binding-predictions>... --output <output>

FLAGS:
        --drop-default    Drop default measures based on TCR and KIR motifs
    -h, --help            Prints help information
    -q, --quiet           Disables any information being printed to terminal (except errors)
        --settings        Lists default measure names and motif positions as well as the default location updated kir
                          ligand will be stored
    -u, --unique          Whether only unique peptide/motif sequences should be considered in the calculations
        --update          Updates the current kir ligand group data
    -V, --version         Prints version information
    -v, --verbose         Determines verbosity of the processing, can be specified multiple times -vvv

OPTIONS:
    -b, --binding-predictions <binding-predictions>...
            Path to files containing predicted Class I affinity data (NetMHCpan results)

    -c, --cohort <cohort>
            Cohort of individuals for which all measures will be calculated

    -i, --index <index>...
            Index allele used for cohort calculations only, all individuals will be compared to these alleles

    -m, --measure <measure>...
            Custom motif positions to use for calculations (format `Name:index,index..` e.g. KIR:2,7,8,9)

    -o, --output <output>                                 Directory to store outputs
    -p, --peptide-length <peptide-length>...
            Which length of input peptide sequence to consider [default: 9]  [possible values: 8, 9, 10, 11]

        --prefix <prefix>                                 Prefix to assign to all outputs
    -t, --threads <threads>                               Number of threads [default: 4]


Example

The following are short examples on a linux/osx based system and for further details and other OS directions see the documentation.

The target/release/fs-tool is for examples after building the tool from source if you downloaded the binary replace it with the path to the binary.

Allele calculations only

To run comparison on positions 1,3,7, and to name the output measure Example while updating data from EBI:

$ ./target/release/fs-tool -b tests/input/binding_predictions/netmhcpan_wBA.txt  --prefix "example_cohort_Gag_180_209" -o example_result

to drop the default measures TCR and KIR the flag --drop-default-measures can be used.

Cohort calculations

To perform calculations, using the default measures, for A02:01 and C08:02 the following command can be ran:

$ ./target/release/fs-tool   -b tests/input/binding_predictions/netmhcpan_wBA.txt  --prefix "example_cohort_Gag_180_209" -o example_result -i A03:01 C08:02 -c tests/input/cohorts/example_cohort.csv

Output

The created directory example_result will contain the following output:

File name Description
example_cohort_Gag_180_209_allele_binding_summary.csv summary of allele peptide binding counts per protein
example_cohort_Gag_180_209_allele_fs_result.csv fraction shared calculation results for all combinations of alleles in the binding predictions
example_cohort_Gag_180_209_allele_metadata.csv lists netmhcpan nearest neighbour information and what allele was the ligand motif assignment based on
example_cohort_Gag_180_209_cohort_result.csv lists per cohort subject calculations for each index allele, peptide length, measure

Authors and Citation

Please cite eLife 2020;9:e54558.

Citation

License

The MIT License (MIT). Please see License File for more information.

About

Tool to calculate fraction of shared bound peptides between HLA proteins

Resources

License

Stars

Watchers

Forks

Packages

No packages published