Multitask learning for host–pathogen protein interactions

Kshirsagar, Meghana; Carbonell, Jaime; Klein-Seetharaman, Judith

There has been little work on combining PPI datasets with the goal of improving prediction performance for multiple organisms. Qi et al. (2010) proposed a semi-supervised multitask framework to predict PPIs from partially labeled reference sets. The basic idea is to perform multitask learning on a supervised classification task and a semi-supervised auxiliary task via a regularization term. Another line of work in PPI prediction (Xu et al., 2010) uses the Collective Matrix Factorization (CMF) approach proposed by Singh and Gordon (2008) . The CMF method learns models for multiple networks by simultaneously factorizing several adjacency matrices and sharing parameters amongst the factors. Xu et al. (2010) use these ideas in their transfer learning setting, where the source network is a relatively dense interaction network of proteins and the objective is to infer PPI edges in a relatively sparse target network. To compute similarities between the nodes in the source and target networks, they use protein sequences and the topological structures of the interaction networks.