So far, comparative tools for exploring the potential influences of species-specific PTMs on host-virus interactions have not been found. Here we develop a web-based

learn more interactive database – CAPIH (Comparative Analysis of Protein Interactions for HIV-1) – for comparative studies of genetic differences between the human proteins involved in host-HIV protein interactions and their orthologues retrieved from three mammalian species: chimpanzee (Pan troglodyte), rhesus macaque (Macaca mulatta), and mouse (Mus musculus). The three latter species are all important animal models for HIV studies [15–17]. Understanding the differences in host-virus interplay between human and the model species is the basis for correct interpretation ALK activation of animal-based HIV studies. Furthermore, by comparing protein interactions between species, one can potentially identify key differences that underlie chimpanzee resistance to AIDS. To facilitate inter-species comparisons of host-HIV PPIs, four main functions are provided in CAPIH. Firstly, the GW-572016 cell line interface shows the presence or absence of orthologous proteins, thus enabling users to pinpoint missing protein components in the host-HIV interaction network.

Secondly, the multiple sequence alignments of orthologous proteins enable users to identify species-specific amino acid substitutions, nucleotide substitutions, and indels. This information is helpful for inferring functional changes of orthologous proteins. Thirdly, predictions of 7 types of species-only PTMs (phosphorylation, methylation, sumoylation, acetylation, sulfation, N-glycosylation, and O-glycosylation) for each HIV-interacting host protein Clomifene are presented for analyses of potential PTM influences on protein interactions and signal/regulatory pathway. We also collect experimentally verified PTMs in human proteins. Fourthly, CAPIH shows potential PPI hot sites on the multiple sequence alignments. Through the visualized interface, researchers can easily spot multiple host factors that directly or indirectly interact

with the same HIV protein, and consider how changes in one member protein may affect the protein interaction network. Construction and content CAPIH organization and implementation The data compiling process is illustrated in Figure 1A. We retrieved a total of 1,447 HIV-1 interacting human proteins from the HIV-1, Human Protein Interaction Database [18] (the November 13, 2007 freeze). The human-chimpanzee-macaque-mouse orthologous proteins were downloaded from the Ensembl genome browser (release 47), which were identified by the Ensembl project using the Markov clustering algorithm [19]. Note that not all the retrieved human proteins have orthologues in all of the three compared species. In the cases of one-to-many/many-to-many orthologous relationships, only the protein pairs with the reciprocally highest similarity were selected. All of the protein and nucleotide sequences were downloaded from Ensembl.