HFADE-FMD: a hybrid approach of fireworks algorithm and differential evolution strategies for functional module detection in protein-protein interaction networks

Functional module detection in protein-protein interaction (PPI) network is one important content of the proteomics research in the post-genomic era. Nowadays the swarm intelligence and evolutionary based approaches have become effective ways for detecting functional modules. This paper proposes a novel hybrid approach of fireworks algorithm and differential evolution strategies for functional module detection in PPI networks (called HFADE-FMD). HFADE-FMD first initializes each firework individual into a candidate functional module partition based on label propagation according to the topological and functional information between protein nodes. Then HFADE-FMD uses the explosion operator of firework algorithm, and mutation, crossover and selection strategies of differential evolution algorithm to iteratively search for better functional module partitions. To verify the performance of HFADE-FMD, this paper compared it with ten competitive methods on four public PPI datasets. The experimental results show that HFADE-FMD achieves prominent performance with respective to Recall, Sn, PPV, and ACC metrics while performing well in terms of Precision and F-measure metrics. Thus, it is able to more accurately detect functional modules and help biologists to find some novel biological insights.

Automated summary

This paper presents a novel hybrid approach of fireworks algorithm and differential evolution strategies for detecting functional modules in PPI networks, which performs well in terms of Recall, Sn, PPV, and ACC metrics. It is able to accurately detect functional modules and assist biologists in discovering novel biological insights.