gpu-coloc
gpu-coloc is an accelerated implementation of the coloc.bf_bf Bayesian colocalization algorithm (Chris Wallace, "A more accurate method for colocalisation analysis allowing for multiple causal variants", PLOS Genetics, 2021, coloc GitHub), delivering identical results to R's coloc.bf_bf approximately 1000 times faster (Also on CPUs!). For GPU-acceleration Torch support is required, thus gpu-coloc can be used on M-series Mac(book)s and machines with NVIDIA GPUs.
If you have any questions or problems with gpu-coloc, please write to mihkel.jesse@gmail.com.
Citation
If you use gpu-coloc, please cite: https://doi.org/10.1101/2025.08.25.672103
All of eQTL Catalogue ready for use with gpu-coloc can be found here.
Installation
Install via pip (Python ≥3.12):
pip install gpu-coloc
If pip cannot find the package, it is probably do to an older version of pip. Thus to upgrade it run:
pip install --upgrade pip
You can set up gpu-coloc up in a virtual environment:
python -m venv [path_to_venv]
source [path_to_venv]/bin/activate
pip install gpu-coloc
This is prefered in production environments or if you are having trouble with package management.
Verify Installation
To confirm installation, clone this repository:
git clone https://github.com/mjesse-github/gpu-coloc
cd gpu-coloc
bash test.sh
This creates an example/ directory containing an example_results.tsv file.
Workflow
Note: Paths assume gpu-coloc is in the working directory; adjust paths if necessary.
Variant Naming Convention
Variants must follow the naming format: chr[chromosome]_[position]_[ref]_[alt]. Ensure renaming is completed before Step 1. Use chromosome X, not 23.
1. Prepare Signals and Summary Files
- Signal Files: Save signals as
[signal].pickle, containing variants and their log Bayes Factors (lbf).
Example format:
variant chrX_153412224_C_A chrX_153412528_C_T ...
lbf -0.060991 -1.508802 ...
- Summary File: Tab-separated, structured as:
signal chromosome location_min location_max signal_strength lead_variant
QTD000141_ENSG00000013563_L1 X 153412224 155341332 12.1069377174147 chrX_154403855_T_G
...
Naming examples:
- Summary:
gwas_summary.tsv - Signals directory:
gwas_signals/[signal].pickle
See scripts in summary_and_signals_examples/ for reference; modifications may be necessary.
2. Format Data
gpu-coloc --format --input [path_to_signals] --input_summary [summary_file] --output [output_folder]
Parameters for gpu-coloc --format
--input(required): Path to the directory containing signal files (e.g.,gwas_signals/). Supportspickleorfeatherfiles (see--input_type).--output(required): Path to the output directory where formatted parquet files will be saved.--input_summary(required): Path to the summary TSV file (e.g.,gwas_summary.tsv).--output_summary(optional): Path to write the formatted summary as a parquet file.--input_type(optional): Type of input files, either'pickle'(default) or'feather'.
3. Run Colocalization
gpu-coloc --run --dir1 [formatted_dataset_1] --dir2 [formatted_dataset_2] --results [results_output] --p12 1e-6 --H4 0.8
Parameters for gpu-coloc --run
--dir1(required): Path to the first formatted dataset directory.--dir2(required): Path to the second formatted dataset directory.--results(required): Path to the output file where colocalization results will be saved.--p12(optional): Prior probability that a variant is associated with both traits. Default:1e-6.--H4(optional): Posterior probability threshold for declaring colocalization (H4). Default:0.8.
1 maintainer
Metadata
1 maintainer
Release files for gpu-coloc 0.7.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | |
|---|---|---|---|---|
| gpu_coloc-0.7.1.tar.gz | 12.6 kB | Without provenance | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| gpu_coloc-0.7.1-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 25.9 kB
Release files / gpu_coloc-0.7.1.tar.gz
| Download URL | gpu_coloc-0.7.1.tar.gz |
|---|---|
| Size | 12.6 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
0b222fca2663f017a314eaac77fe3dea008ec571ea3aba4c54eb966941614c6f
|
|
BLAKE2b-256 checksum How to use checksums |
3a213751f56f62f4806a8e588c92a48edd41871a3329e40c8a84a7aaf8072ee1
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.1.0 CPython/3.12.3
|
Release files / gpu_coloc-0.7.1-py3-none-any.whl
| Download URL | gpu_coloc-0.7.1-py3-none-any.whl |
|---|---|
| Size | 13.4 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
6a39f5086c2040085c340c3e7b4244358e0c24f4c30ba0b7616444a15c91ecfe
|
|
BLAKE2b-256 checksum How to use checksums |
4bd5408e75eb848b09ce7cd00415bbb8fd0feba49a2bb560385dd882a6524dbf
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.1.0 CPython/3.12.3
|
Security for gpu-coloc 0.7.1
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.
