Inference of Gene Regulatory Networks Based on Nonlinear Ordinary Differential Equations

Baoshan Ma1,*, Mingkun Fang1 and Xiangtian Jiao1

1 College of Information Science and Technology, Dalian Maritime University, Dalian 116039, China

The proposed method is a scalable method exploiting time-series and steady-state data jointly, in which nonlinear ODEs and XGBoost are employed to infer gene regulatory networks.

If you find our method is useful, please cite our paper: Baoshan Ma, Mingkun Fang, Xiangtian Jiao. Inference of gene regulatory networks based on nonlinear ordinary differential equations. Bioinformatics, 2020,36(19):4885-4893. https://doi.org/10.1093/bioinformatics/btaa032

The describe of the program

The program of xgbgrn can combine time-series data and steady-state data to infer GRNs, the steady-state data is not necessary.

The program of xgbgrn_2 can only be applied to one type of data.

The version of Python and packages

Python version=3.6
Xgboost version=0.82
scikit-learn version=l0.24.2
numpy version=1.16.3

Parameters

xgbgrn:
    TS_data: a matrix of time-series data
    time_points: a list of time points
    alpha:a constant or specify "from_data"
    SS_data: a matrix of time-series data, the default is "none"
    gene_names: a list of gene names
    regulators: a list of names of regulatory genes, the default is "all", 
    param: a dict of parameters of xgboost
	
xgbgrn_2:
    expr_data: a matrix of gene expression data
    gene_names: a list of gene names
    regulators: a list of names of regulatory genes
    param: a dict of parameters of xgboost