Heavy path mining of protein–protein associations in the malaria parasite

• We develop a heavy path network-mining algorithm for protein association networks.
• Using it, we predict potential central players in the malaria parasite life cycle.
• This method can be extended to other organisms for network mining and annotation.

Annotating and understanding the function of proteins and other elements in a genome can be difficult in the absence of a well-studied and evolutionarily close relative. The causative agent of malaria, one of the oldest and most deadly global infectious diseases, is a good example of this problem. The burden of malaria is huge and there is a pressing need for new, more effective antimalarial strategies. However, techniques such as homology-dependent annotation transfer are severely impaired in this parasite because there are no well-understood close relatives. To circumvent this approach we developed a network-based method that uses a heavy path network-mining algorithm. We uncovered the protein–protein associations that are implicated in important cellular processes including genome integrity, DNA repair, transcriptional regulation, invasion, and pathogenesis, thus demonstrating the utility of this method.The URL of the source code for super-sequence mining method is http://www.cs.utsa.edu/~korkmaz/research/heavy-path-mining/.