POP-TOOLS (POst-Prediction TOOLS) is a Python3-based command line toolkit for conducting valid and powerful machine learning (ML)-assisted genetic association studies.
The POP-TOOLS toolkit can be used to conduct
POP-GWAS[POst-Prediction Genome-wide Association Studies (GWAS)]
The POP-TOOLS and its required modules can be installed via
git clone https://github.com/qlu-lab/POP-TOOLS
cd POP-TOOLS
pip install -r requirements.txt
Please see the TL;DR to conduct POP-GWAS
Please see the wiki for tutorials describing the basic function and along with detailed manual of POP-TOOLS.
Please see the FAQ for frequently asked questions related to POP-TOOLS.
We provide a web interface for the power and sample size calculation for ML-assisted GWAS.
[Version 1.1.0] (May 1, 2024): Added quality control to remove SNPs with duplicate IDs; Added a version of the sample overlap correction; Modified scipts to accommodate the latest version of polars.
[Version 1.0.0] (Jan 2, 2024): Initial release.
Valid inference for machine learning-assisted GWAS
Assumption-lean and Data-adaptive Post-Prediction Inference
For questions and comments, please open a GitHub issue (preferred) or contact Jiacheng Miao at jiacheng.miao@wisc.edu or Qiongshi Lu at qlu@biostat.wisc.edu.
- POPInf (POst-Prediction Inference) is a generic toolkit for conducting valid and powerful post-prediction inference. It is more general (can be applied to a wider range of statistical quantities), but is not optimized for genetic applications.