BIPS: BIANA Interolog Prediction Server. A tool for protein–protein interaction inference

Structural Bioinformatics Laboratory (GRIB-IMIM), Universitat Pompeu Fabra, Barcelona Research Park of Biomedicine (PRBB), 08003 Barcelona, Catalonia, Spain.
Nucleic Acids Research (Impact Factor: 9.11). 06/2012; 40(Web Server issue):W147-51. DOI: 10.1093/nar/gks553
Source: PubMed


Protein-protein interactions (PPIs) play a crucial role in biology, and high-throughput experiments have greatly increased the coverage of known interactions. Still, identification of complete inter- and intraspecies interactomes is far from being complete. Experimental data can be complemented by the prediction of PPIs within an organism or between two organisms based on the known interactions of the orthologous genes of other organisms (interologs). Here, we present the BIANA (Biologic Interactions and Network Analysis) Interolog Prediction Server (BIPS), which offers a web-based interface to facilitate PPI predictions based on interolog information. BIPS benefits from the capabilities of the framework BIANA to integrate the several PPI-related databases. Additional metadata can be used to improve the reliability of the predicted interactions. Sensitivity and specificity of the server have been calculated using known PPIs from different interactomes using a leave-one-out approach. The specificity is between 72 and 98%, whereas sensitivity varies between 1 and 59%, depending on the sequence identity cut-off used to calculate similarities between sequences. BIPS is freely accessible at

Download full-text


Available from: Baldomero Oliva,
  • Source
    • "A total of 631 933 positive pairs and 13 543 217 negative pairs were used for assessing giant panda networks, and 463 223 positive pairs and 13 111 432 negative pairs were used for assessing soybean networks. We compared the quality of the orthology-based networks by JiffyNet and BIPS, another public web server capable of constructing orthology-based protein networks (10), using identical query protein sequence input data. The BIPS web server differs substantially from JiffyNet as described above. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Revolutionary DNA sequencing technology has enabled affordable genome sequencing for numerous species. Thousands of species already have completely decoded genomes, and tens of thousands more are in progress. Naturally, parallel expansion of the functional parts list library is anticipated, yet genome-level understanding of function also requires maps of functional relationships, such as functional protein networks. Such networks have been constructed for many sequenced species including common model organisms. Nevertheless, the majority of species with sequenced genomes still have no protein network models available. Moreover, biologists might want to obtain protein networks for their species of interest on completion of the genome projects. Therefore, there is high demand for accessible means to automatically construct genome-scale protein networks based on sequence information from genome projects only. Here, we present a public web server, JiffyNet, specifically designed to instantly construct genome-scale protein networks based on associalogs (functional associations transferred from a template network by orthology) for a query species with only protein sequences provided. Assessment of the networks by JiffyNet demonstrated generally high predictive ability for pathway annotations. Furthermore, JiffyNet provides network visualization and analysis pages for wide variety of molecular concepts to facilitate network-guided hypothesis generation. JiffyNet is freely accessible at
    Nucleic Acids Research 05/2013; 41(Web Server issue). DOI:10.1093/nar/gkt419 · 9.11 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Protein-protein interactions play a relevant role among the different functions of a cell. Identifying the protein-protein interaction network of a given organism (interactome) is useful to shed light on the key molecular mechanisms within a biological system. In this work, we show the role of structural features (loops and domains) to comprehend the molecular mechanisms of protein-protein interactions. A paradox in protein-protein binding is to explain how the unbound proteins of a binary complex recognize each other among a large population within a cell and how they find their best docking interface in a short time-scale. We use interacting and non-interacting protein pairs to classify the structural features that sustain the binding (or non-binding) behaviour. Our study indicates that not only the interacting region but also the rest of the protein surface is important for the interaction fate. The interpretation of this classification suggests that the balance between favouring and disfavouring structural features determines if a pair of proteins interacts or not. Our results are in agreement with previous works and support the funnel-like intermolecular energy landscape theory that explains protein-protein interactions. We have used these features to score the likelihood of the interaction between two proteins and to develop a method for the prediction of PPIs. We have tested our method on several sets with unbalanced ratios of interactions and non-interactions to simulate real conditions, obtaining accuracies higher than 25% in the most unfavourable circumstances.
    Journal of Molecular Biology 01/2013; 425(7). DOI:10.1016/j.jmb.2013.01.014 · 4.33 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Protein interaction networks are important for the understanding of regulatory mechanisms, for the explanation of experimental data and for the prediction of protein functions. Unfortunately, most interaction data is available only for model organisms. As a possible remedy, the transfer of interactions to organisms of interest is common practice, but it is not clear when interactions can be transferred from one organism to another and, thus, the confidence in the derived interactions is low. Here, we propose to use a rich set of features to train Random Forests in order to score transferred interactions. We evaluated the transfer from a range of eukaryotic organisms to S. cerevisiae using orthologs. Directly transferred interactions to S. cerevisiae are on average only 24% consistent with the current S. cerevisiae interaction network. By using commonly applied filter approaches the transfer precision can be improved, but at the cost of a large decrease in the number of transferred interactions. Our Random Forest approach uses various features derived from both the target and the source network as well as the ortholog annotations to assign confidence values to transferred interactions. Thereby, we could increase the average transfer consistency to 85%, while still transferring almost 70% of all correctly transferable interactions. We tested our approach for the transfer of interactions to other species and showed that our approach outperforms competing methods for the transfer of interactions to species where no experimental knowledge is available. Finally, we applied our predictor to score transferred interactions to 83 targets species and we were able to extend the available interactome of B. taurus, M. musculus and G. gallus with over 40,000 interactions each. Our transferred interaction networks are publicly available via our web interface, which allows to inspect and download transferred interaction sets of different sizes, for various species, and at specified expected precision levels. Availability:
    PLoS ONE 06/2013; 8(6):e66635. DOI:10.1371/journal.pone.0066635 · 3.23 Impact Factor
Show more