gpuZoo

Reference

gpuZoo: Cost-effective estimation of gene regulatory networks using the Graphics Processing Unit.

Abstract

Gene regulatory network inference allows the study of transcriptional control to identify the alteration of cellular processes in human diseases. The computational inference of gene regulation using PANDA, PUMA, and SPIDER for aggregate networks and LIONESS, which models single-sample networks involves the assessment of several lines of biological evidence. These tools use the same PANDA computational backend, in particular, they perform operations on large data matrices, which limits their use for large-scale genomic studies due to the computational burden. To address this limitation, we developed gpuZoo, a GPU-accelerated implementation of PANDA, PUMA, SPIDER, and LIONESS. The runtime of the gpuZoo implementation in MATLAB and Python is up to 35 times faster and 20 times less expensive than multi-core CPU implementations of the same method. gpuZoo takes advantage of the modern multi-GPU device architecture to build a population of sample-specific gene regulatory networks with similar runtime and cost improvements. Taken together, gpuZoo allows gene regulatory network inference in large-scale genomic studies with cost-effective performance.

gpuZoo is available in MATLAB through the netZooM package https://github.com/netZoo/netZooM and in Python through the netZooPy package https://github.com/netZoo/netZooPy.

Supplementary data

This is the data used for the benchmarks of gpuZoo.

Model Motif PPI Expression Mode TFs Genes
Small network download download download Intersection 652 1000
Protein-coding genes network download download download Union 652 27149
Transcript network download download download Union 1603 43698

Implementation

netZooPy, netZooM

GitHub repository

To reproduce the benchmarks, check the github repository gpuZoo.