The Structural Bioinformatics Research Group led by Baldo Oliva, a researcher in UPF's Department of Experimental and Health Sciences (CEXS), has presented two web servers that will facilitate the study of protein configuration and the gene expression involved in the manifestation of several conditions.

One of the servers developed by the research group in Structural Bioinformatics, named FRAGRUS, is focused on protein design. This work has been published in the journal Bioinformatics and has been undertaken in collaboration with Narcís Fernández-Fuentes, head of the Bioinformatics group at the University of Aberystwyth (UK), who is doing a research stay at Universitat Pompeu Fabra for two years through an ACCIÓ TecnioSpring grant.

In computational design based on protein structure, it is common to remodel one or more short fragments of the protein's structure to achieve new functions or new interactions. This remodelling can be new, or it can be the result of the redistribution of already existing fragments, extracted from known protein structures.

FRAGRUS is a web server designed to test alternative conformations of proteins extracted from fragments of functional super secondary structures, taken from known protein structures. The method uses a database of secondary structural motifs called smotifs.

GUILDify: a web server for the phenotypic characterization of genes

Protein interaction networks, whether or not their structure is known, are valuable tools for the study of medical conditions. In recent years, the prioritization of genes associated to conditions has been crucial in the identification of new therapeutic targets, especially in cases where multiple genes act cooperatively, hampering the search.

According to Baldo Oliva, "the study of physical interactions between proteins encoded by genes involved in specific conditions has enabled the use of the principle of "guilt by association" to predict these new targets."

The GUILDify web server is easy to use for the phenotypic characterization of genes and it is presented in an article that has also been published in the journal Bioinformatics. This server is expected to be useful for studying conditions, predicting therapeutic targets, and suggesting the most appropriate drugs for any specific condition selected by the user.

"The GUILDify web server does not restrict the assignment of priorities to any predefined phenotype, allowing the search of genes specified by the user. It also prioritizes drugs depending on their therapeutic targets, although reusing the same drugs as potential targets for new therapies is not ruled out," says Baldo Oliva.

References:

Jaume Bonet, Joan Segura, Joan Planas-Iglesias, Baldomero Oliva i Narcís Fernández-Fuentes (2014), "Frag'rUs: knowledge-based sampling of protein backbone conformations for de novo structure-based protein design", Bioinformatics, doi: 10.1093/bioinformatics/btu129.

Emre Guney, Javier García-García, Baldo Oliva (2014), "GUILDify: a web server for phenotypic characterization of genes through biological data integration and network-based priorization algorithms", Bioinformatics, doi: 10-1093/bioinformatics/btu092.