idc9 / jackstraw

implements jackstraw methods in python test for associations between variables and systematic modes of variation discovered in the data.

jackstraw

author: Iain Carmichael

Additional documentation, examples and code revisions are coming soon. For questions, issues or feature requests please reach out to Iain: [e-mail].

Overview

This package performs association tests between the observed data and their systematic patterns of variation. This package implements methods from (Chung and Storey, 2015) in python. For an R version (which we followed closely) see https://github.com/ncchung/jackstraw.

Chung, N.C. and Storey, J.D. (2015) Statistical significance of variables driving systematic variation in high-dimensional data. Bioinformatics, 31(4): 545-554 http://bioinformatics.oxfordjournals.org/content/31/4/545

Installation

This is currently an informal package under development so I've only made it installable from github.

git clone https://github.com/idc9/jackstraw.git
python setup.py install

Example

import numpy as np
from jackstraw.jackstraw import Jackstraw

X = np.random.normal(size=(100, 20))
jack = Jackstraw(S = 10, B = 100)
jack.fit(X, method='pca', rank=4)
jack.rejected

array([], dtype=int64)

For some more example code see these example notebooks.

Help and Support

Additional documentation, examples and code revisions are coming soon. For questions, issues or feature requests please reach out to Iain: [e-mail].

Documentation

The source code is located on github: https://github.com/idc9/jackstraw.

Testing

Testing is done using nose.

Contributing

We welcome contributions to make this a stronger package: data examples, bug fixes, spelling errors, new features, etc.