scSIGRN 0.0.7

GRN inference

SIGRN

SIGRN: Inferring Gene Regulatory Network with Soft Introspective Variational Autoencoders

Architecture

Dependencies

• python =3.8
• torch==2.1.0
• scanpy==1.9.1
• other detailed installation packages can be found in requirements.txt
• CUDA toolkit 11.0 or later.

Installation

pip install scSIGRN==0.0.7

Data Preparation

In our study, we trained our model using data from BEENLINE. You can download the datasets from the provided link.

Example tutorial

We provide an example tutorial. Check out the this tutorial for a quick overview of how to use SIGRN for your research!

Usage

SIGRN accepts input data in CSV, TSV format, or H5AD format as provided by Scanpy (genes in rows and cells in columns for TSV and CSV). The output of the GRN inference task includes an adjacency matrix and various evaluation metrics, such as AUC, EPR, and AUPRR.

Baseline methods

• Beeline https://github.com/Murali-group/Beeline/tree/master
• DeepSEM https://github.com/HantaoShu/DeepSEM
• GRN-VAE/DAZZLE https://github.com/TuftsBCB/dazzle/tree/main

References

• Thanks to the following authors for their papers and codes.

  [1] Pratapa, A., Jalihal, A.P., Law, J.N., Bharadwaj, A., Murali, T.: Benchmarking algorithms for gene regulatory network inference from single-cell transcriptomic data. Nature methods 17(2), 147–154 (2020)

  [2] Shu, H., Zhou, J., Lian, Q., Li, H., Zhao, D., Zeng, J., Ma, J.: Modeling gene regulatory networks using neural network architectures. Nature Computational Science 1(7), 491–501 (2021)

  [3] Zhu, H., Slonim, D.: Grn-vae: A simplified and stabilized sem model for gene regulatory network inference. bioRxiv pp. 2023–01 (2023)