Prediction of interactions between HIV-1 and human proteins by information integration

Abstract

Human immunodeficiency virus-1 (HIV-1) in acquired immune deficiency syndrome (AIDS) relies on human host cell proteins in virtually every aspect of its life cycle. Knowledge of the set of interacting human and viral proteins would greatly contribute to our understanding of the mechanisms of infection and subsequently to the design of new therapeutic approaches. This work is the first attempt to predict the global set of interactions between HIV-1 and human host cellular proteins. We propose a supervised learning framework, where multiple information data sources are utilized, including co-occurrence of functional motifs and their interaction domains and protein classes, gene ontology annotations, posttranslational modifications, tissue distributions and gene expression profiles, topological properties of the human protein in the interaction network and the similarity of HIV-1 proteins to human proteins' known binding partners. We trained and tested a Random Forest (RF) classifier with this extensive feature set. The model's predictions achieved an average Mean Average Precision (MAP) score of 23%. Among the predicted interactions was for example the pair, HIV-1 protein tat and human vitamin D receptor. This interaction had recently been independently validated experimentally. The rank-ordered lists of predicted interacting pairs are a rich source for generating biological hypotheses. Amongst the novel predictions, transcription regulator activity, immune system process and macromolecular complex were the top most significant molecular function, process and cellular compartments, respectively. Supplementary material is available at URL www.cs.cmu.edu/õznur/hiv/hivPPI.html

Figure 1

Schematic showing features that incorporate knowledge of the human protein interactome. These features can be conceptually grouped into two categories: 1) graph properties of human protein j in human protein interaction network, which include degree, clustering coefficient and betweenness centrality of node j 2) the similarity of the HIV-1 protein, i, to human protein j’s interaction partners denoted by f_neigh (i,j) in the figure. The maximal similarity to the neighbors is used. Five features are derived; GO function, process and location similarity in addition to post translational modification and sequence similarity.

Figure 2

The average precision vs. recall curve of the Random Forest model trained on the complete feature set, in comparison to models trained with a subset of features. The top 3 Gini features are degree, betweenness centrality, and GO neighbor process similarity features. The top 6 Gini features are the top 3 Gini features plus clustering coefficient, GO neighbor function, and location features. These are compared to two baseline classifiers, where 6 features are randomly selected from the set of features that does not include the top 6 Gini features, with and without protein type features (ptf).

Figure 3

RF Gini importance measures for each feature. Protein type features are grouped together.