Master status:

Development status:

Package information:

A scikit-learn-compatible Python implementation of Multifactor Dimensionality Reduction (MDR) for feature construction. This project is still under active development and we encourage you to check back on this repository regularly for updates.

MDR is an effective feature construction algorithm that is capable of modeling higher-order interactions and capturing complex patterns in data sets.

MDR currently only works with categorical features and supports both binary classification and regression problems. We are working on expanding the algorithm to cover more problem types and provide more convenience features.

Please see the repository license for the licensing and usage information for the MDR package.

Generally, we have licensed the MDR package to make it as widely usable as possible.

MDR is built on top of the following existing Python packages:

• NumPy

• SciPy

• scikit-learn

• matplotlib

All of the necessary Python packages can be installed via the Anaconda Python distribution, which we strongly recommend that you use. We also strongly recommend that you use Python 3 over Python 2 if you're given the choice.

NumPy, SciPy, scikit-learn, and matplotlib can be installed in Anaconda via the command:

conda install numpy scipy scikit-learn matplotlib

Once the prerequisites are installed, you should be able to install MDR with a pip command:

pip install scikit-mdr

Please file a new issue if you run into installation problems.

MDR has been coded with a scikit-learn-like interface to be easy to use. The typical fit, transform, and fit_transform methods are available for every feature construction algorithm. For example, MDR can be used to construct a new feature composed from two existing features:

from mdr import MDR
import pandas as pd

genetic_data = pd.read_csv('https://github.com/EpistasisLab/scikit-mdr/raw/development/data/GAMETES_Epistasis_2-Way_20atts_0.4H_EDM-1_1.tsv.gz', sep='\t', compression='gzip')

features = genetic_data.drop('class', axis=1).values
labels = genetic_data['class'].values

my_mdr = MDR()
my_mdr.fit(features, labels)
my_mdr.transform(features)
>>>array([[1],
>>>       [1],
>>>       [1],
>>>       ...,
>>>       [0],
>>>       [0],
>>>       [0]])

You can also use MDR as a classifier, and evaluate the quality of the constructed feature with the score function:

from mdr import MDRClassifier
import pandas as pd

genetic_data = pd.read_csv('https://github.com/EpistasisLab/scikit-mdr/raw/development/data/GAMETES_Epistasis_2-Way_20atts_0.4H_EDM-1_1.tsv.gz', sep='\t', compression='gzip')

features = genetic_data.drop('class', axis=1).values
labels = genetic_data['class'].values

my_mdr = MDRClassifier()
my_mdr.fit(features, labels)
my_mdr.score(features, labels)
>>>0.998125

If you want to use MDR for regression problems, use ContinuousMDR:

from mdr import ContinuousMDR
import pandas as pd

genetic_data = pd.read_csv('https://github.com/EpistasisLab/scikit-mdr/raw/development/data/GAMETES_Epistasis_2-Way_continuous_endpoint_a_20s_1600her_0.4__maf_0.2_EDM-2_01.tsv.gz', sep='\t', compression='gzip')
features = genetic_data[['M0P0', 'M0P1']].values
targets = genetic_data['Class'].values

my_cmdr = ContinuousMDR()
my_cmdr.fit(features, targets)
my_cmdr.transform(features)
>>>array([[0],
>>>       [1],
>>>       [1],
>>>       ...,
>>>       [0],
>>>       [1],
>>>       [1]])

We welcome you to check the existing issues for bugs or enhancements to work on. If you have an idea for an extension to the MDR package, please file a new issue so we can discuss it.

Please check the existing open and closed issues to see if your issue has already been attended to. If it hasn't, file a new issue on this repository so we can review your issue.

If you use this software in a publication, please consider citing it. You can cite the repository directly with the following DOI:

[blank for now]

The MDR package was developed in the Computational Genetics Lab with funding from the NIH. We're incredibly grateful for their support during the development of this project.