Skip to content

XSLiuLab/MethyDeep

Repository files navigation

Accurate prediction of pan-cancer types using machine learning with minimal number of DNA methylation sites

License version

Table of content

Table of content


Overview

This repository provides the analysis reports, code and data for readers who are interest in this project and make it easier to reproduce the whole analysis procedure.

Contents

Brief use

If you just want to use our model to predict, follow the following process

### load model
from keras.models import load_model
import pandas as pd
import numpy as np
new_model = load_model('./MethyDeep')
feature = pd.read_csv("./data/RF_feature45.csv")
target_names = pd.read_csv("./data/targer_names.csv")
### load data
data = pd.read_csv("./yourdata")
data = data[feature.iloc[0:30,0]]
#### predict
predict = np.argmax(new_model.predict(data), axis=-1)
predict2 = []
for j in predict: 
    predict2.append(target_names[j])
predict2 ### predict result

Citation

Acknowledgement

We thank ShanghaiTech University High Performance Computing Public Service Platform for computing services.This work was supported by Shanghai Science and Technology Commission (21ZR1442400), the National Natural Science Foundation of China (31771373), and startup funding from ShanghaiTech University.

License

Ning W, Wu T, Wu C, Wang S, Tao Z, Wang G, Zhao X, Diao K, Wang J, Chen J, Chen F, Liu XS. Accurate prediction of pan-cancer types using machine learning with minimal number of DNA methylation sites. J Mol Cell Biol. 2023 Apr 10:mjad023. doi: 10.1093/jmcb/mjad023. Epub ahead of print. PMID: 37037781.


Cancer Biology Group @ShanghaiTech

Research group led by Xue-Song Liu in ShanghaiTech University

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published