Metadata-Version: 2.4
Name: genboostgpu
Version: 0.1.1
Summary: GENBoostGPU provides a scalable framework for running elastic net regression with boosting across thousands of CpG sites, leveraging GPU acceleration with RAPIDS cuML, CuPy, and cuDF.
License: GPL-3.0-or-later
Keywords: genomics,epigenomics,dna-methylation,elastic-net,boosting,gpu,snp-heritability
Author: Kynon J Benjamin
Author-email: kj.benjamin90@gmail.com
Maintainer: Kynon J Benjamin
Maintainer-email: kj.benjamin90@gmail.com
Requires-Python: >=3.10,<4.0
Classifier: License :: OSI Approved :: GNU General Public License v3 or later (GPLv3+)
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Classifier: Programming Language :: Python :: 3.14
Requires-Dist: cudf-cu12 (>=25.8.0,<26.0.0)
Requires-Dist: cuml-cu12 (>=25.8.0,<26.0.0)
Requires-Dist: cupy-cuda12x (>=13.3.0,<14.0.0)
Requires-Dist: dask-cuda (>=25.8.0,<26.0.0)
Requires-Dist: numba (<0.62)
Requires-Dist: numpy (<2.3.0)
Requires-Dist: optuna (>=4.5.0,<5.0.0)
Requires-Dist: pandas (>=2.3.2,<3.0.0)
Requires-Dist: pandas-plink (>=2.3.2,<3.0.0)
Requires-Dist: scikit-learn (>=1.7.2,<2.0.0)
Project-URL: Bug Tracker, https://github.com/heart-gen/GENBoostGPU/issues
Project-URL: homepage, https://github.com/heart-gen/GENBoostGPU
Project-URL: repository, https://github.com/heart-gen/GENBoostGPU
Description-Content-Type: text/markdown

# GENBoostGPU

**Genomic Elastic Net Boosting on GPU (GENBoostGPU)**

GENBoostGPU provides a scalable framework for running elastic net regression with 
boosting across thousands of CpG sites, leveraging GPU acceleration with RAPIDS cuML, 
CuPy, and cuDF. It supports SNP preprocessing, cis-window filtering, LD clumping, 
missing data imputation, and phenotype integration — all optimized for large-scale 
epigenomics.

---

## Features
- GPU-accelerated **elastic net regression** with optional boosting
- SNP-level preprocessing:
  - Zero-variance SNP filtering
  - Missing genotype imputation
  - LD clumping (PLINK-like) on GPU
- Cis-window filtering for CpGs
- Integration of genotype (PLINK) and phenotype (CpG/VMR methylation) data
- Batch execution across **thousands of CpGs on a single GPU**
- Flexible output: betas, heritability estimates, cross-validation results

---


