Skip to main content

pandas-plink-ng

[!IMPORTANT] This repository is a maintained fork of pandas-plink by Danilo Horta and contributors.
All credit for the original implementation belongs to the upstream authors.
This fork is maintained by Kynon J M Benjamin for compatibility with modern Python and GPU-based genomics workflows.

Upstream project: https://github.com/limix/pandas-plink
Current fork: https://github.com/KrotosBenjamin/pandas-plink-ng

What’s different?

  • Relaxed NumPy dependency → now supports NumPy ≥ 2.0.
  • No other functional changes to core code or API behavior.

Overview

pandas-plink-ng provides Pythonic access to PLINK binary genotype files (.bed/.bim/.fam)
and related matrices (e.g., GCTA .rel files).
It uses lazy loading to minimize memory usage, reading genotypes only when accessed.

Notable historical changes prior to this fork are tracked in the
upstream CHANGELOG.


Installation

Install from PyPI:

pip install pandas-plink-ng

This fork ensures compatibility with the latest RAPIDS, CuPy, and localQTL GPU environments using NumPy 2.x.

If you prefer the original, CPU-focused package:

pip install pandas-plink

Usage

Usage remains identical to the upstream API:

from pandas_plink import read_plink1_bin

G = read_plink1_bin("chr11.bed", "chr11.bim", "chr11.fam", verbose=False)
print(G)

Output (example):

<xarray.DataArray 'genotype' (sample: 14, variant: 779)>
dask.array<shape=(14, 779), dtype=float64, chunksize=(14, 779)>
Coordinates:
  * sample   (sample) object 'B001' 'B002' ... 'B014'
  * variant  (variant) object '11_316849996' ... '11_345698259'
    a0       (variant) <U1 ...
    a1       (variant) <U1 ...

You can also read realized relationship matrices:

from pandas_plink import read_rel

K = read_rel("plink2.rel.bin")
print(K)

For full API details, refer to the original pandas-plink documentation.


Attribution

Original Author: Danilo Horta Upstream License: MIT Maintainer of this fork: Kynon J M Benjamin


License

This project remains under the MIT License. All original copyright notices are retained.

Metadata

Release files for pandas-plink-ng 2.3.3

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for pandas-plink-ng 2.3.3
File Size Uploaded
pandas_plink_ng-2.3.3.tar.gz 25.3 kB Details

Built distribution (wheel)

Table of built distributions (wheels) for pandas-plink-ng 2.3.3
File Interpreter ABI Platform
pandas_plink_ng-2.3.3-cp311-cp311-manylinux_2_28_x86_64.whl CPython 3.11 CPython 3.11 Linux glibc 2.28+ x86-64 Details

Total release size: 60.3 kB

Release history Release notifications | RSS feed

This release

2.3.3 This release

2 release files

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page