xzhang2016 / DeepHE

A deep learning framework for essential gene prediction
MIT License
9 stars 5 forks source link

DeepHE

DeepHE is a deep learning framework for essential gene prediction based on sequence data and PPI network.

Requirements

The code is based on python 3. It depends on several python packages, such as numpy, keras, scikit-learn, networkx, pandas, tensorflow or theano, node2vec, gensim.

Usage

For network features, please follow the instructions of node2vec(or here ).

Command line usage:

$python main.py [--expName EXPNAME] [--fold FOLD] [--embedF EMBEDF]
         [--data_dir DATA_DIR] [--trainProp TRAINPROP] [--repeat REPEAT]
         [--result_dir RESULT_DIR] [--numHiddenLayer NUMHIDDENLAYER]

For more information about the parameters, you can type:

$python main.py --help

or

$python main.py -h