Inference of Large-scale Time-delayed Gene Regulatory Network with Parallel MapReduce Cloud Platform

Bin Yang,Yuehui Chen,Wenzheng Bao,De-Shuang Huang

doi:10.1038/s41598-018-36180-y

Abstract

Inference of gene regulatory network (GRN) is crucial to understand intracellular physiological activity and function of biology. The identification of large-scale GRN has been a difficult and hot topic of system biology in recent years. In order to reduce the computation load for large-scale GRN identification, a parallel algorithm based on restricted gene expression programming (RGEP), namely MPRGEP, is proposed to infer instantaneous and time-delayed regulatory relationships between transcription factors and target genes. In MPRGEP, the structure and parameters of time-delayed S-system (TDSS) model are encoded into one chromosome. An original hybrid optimization approach based on genetic algorithm (GA) and gene expression programming (GEP) is proposed to optimize TDSS model with MapReduce framework. Time-delayed GRNs (TDGRN) with hundreds of genes are utilized to test the performance of MPRGEP. The experiment results reveal that MPRGEP could infer more accurately gene regulatory network than other state-of-art methods, and obtain the convincing speedup.

Highlights

Inferring gene regulatory network (GRN) is the primary and important biochemical network, which contains the regulatory relationships among genes, proteins and small molecules[1,2]
In Chowdhury’s method, differential evolution (DE) algorithm was utilized to optimize all parameters in a time-delayed S-System (TDSS) model, and the computing load is very large for the large-scale GRN
Kimura’s method (S-system model based on decomposition strategy and a cooperative coevolutionary algorithm)[21], DBN22 and TDSS23 are applied for 30-gene artificial Time-delayed GRNs (TDGRN) identification

Summary

Where is the expression level of gene

Xj at t τgij time point, τgij and τhij are the time-delayed factors, N is the total number of genes in TDGRN, αi and βi are rate constants of production function and consumption function, gij and hij are kinetic orders. To infer large-scale gene regulatory network and reduce high computation load, our hybrid evolutionary method based on Hadoop MapReduce framework is proposed. This framework distributes evolutionary tasks to Map and Reduce modules. In order to realize the crossover operation between chromosomes, we randomly divide the population into different partitions. The gained sub offsprings and fitness values are written to output file of the Reduce phase in order to update the input data on the HDFS. If the number of iterations reaches the termination condition, the algorithm is terminated; otherwise, go to the Map phase

Experiments

MPRGEP with computing cluster

Discussion and Conclusion

Additional Information