Some research projects explained

The main research lines of our group are

  • to delineate the emergence and evolution of signaling pathways 
  • to understand mechanisms of protein function in these pathways
  • to identify the regulatory and transcritpional regulation features of these pathways
  • to estimate the impact of alterations of particular components on disease development.


To this purpose we apply systematic analyses using a wide range of bioinformatics and computational tools. From our global analyses we extract useful information that can be narrowed to study particular genes/proteins. The expertise available in the group includes structural bioinformatics, sequence and evolutionary analyses, transcriptional regulation and biological networks.

Evolutionary and mechanistic aspects of the RAS superfamily.

Within the cell, signaling is an essential process that allows cells to quickly adapt to new conditions. Signaling pathways that have been the focus of extensive studies involve RAS proteins as main actors. The hallmark of this particular pathway is the existence of a great and, at the same time subtle, functional diversification in the context of a preserved structural framework.

The RAS super-family of proteins is fundamental to cell organization and signaling, where minor modifications in sequence, structure, and/or cellular regulation will affect binding of regulators. We have conducted computational addressing of the members of the super-family under a comparative genomics framework using a representative set of different species indicating important evolutionary times (Rojas et al, 2012; Rojas and Valencia, 2014).

Our analyses permitted to provide an evolutionary view of the emergence of these proteins and therefore reclassify the human RAS catalogue, identified the founder members (Figure C) and clarified the position of certain human proteins, difficult to classify, in the tree. We have also analyzed on depth the particular “sub-functionalization code” in the G-domain (Figure D) and provided a structural framework for those specificities (Figure E).

In addition, by a combination of experimental and computational analyses we have determined that CENP-M is a pseudo-GTPase which drives the human kinetochore assembly (Basilisco et al., 2014). The relationship was very difficult to establish using standard searches due to the fact that CENPM lacks the switch I region. To determine the relationship we used extensive profile and structure-based analyses. 


We are currently investigating the transcriptional features of this signaling pathway. 

Comparative analyses of DNA Damage Response components and their impact on disease

The DNA Damage Response is a compendium of distinct but frequently overlapping pathways essential for cell viability. They include, sensing of the damage, checking the damage, and repair or die. Many proteins are recruited to and from the damage sites and this recruitment/depletion is regulated by post-translational modifications.

Given the imporance of this pathway, one would expect to find a large degree of conservation along the evolutionary scale. Interestingly, so far there is not a clear picture how this pathway has emerged, nor how its fundamental features have shaped the different sub-networks. Using manually curated information from H. sapiens, we have computationally approached its emergence, the evolution of certain components using large-scale phylogenetic approaches, and have addressed the evolution of the post-translational modifications involving pairs of interactors involved in different aspects of the pathway (Arcas et al, Mol. Biol. Evol, 2014).

We have compiled the most complete network of human DDR pathways including regulatory aspects, and studied its emergence within a global evolutionary framework. The vast majority of these components have an ancient origin and while it is not surprising that the metabolic components of the network were predominant at early evolutionary times, so were the regulatory activities, even though they have subsequently expanded steadily during evolution. Repair based on the NHEJ pathway is probably the oldest part of the network, where similarities in prokaryotes can only be detected using sensitive structure-based methods, and where both canonical and non-canonical pathways are present. The newest acquisition is the response to DSB mediated by ATM, which seems to have grown by assembling existing components (i.e.: the BRCA1-module) and including post-translational modifications that affect protein complexes coupled to the regulation of the cell cycle. Entire pathways have been lost in some model organisms, and remarkable gene loses was observed in invertebrates. Moreover, gene loss in regulatory modules could have influenced the regulation of DDR in entire lineages (i.e.: Nematoda vs. Annelida), where additional compensatory systems may have taken over.


Emergence and evolution of signaling modules

The emergence of signaling modules poses a long standing quesiton in Biology. A widely used approach is to account for the presence of protein domains known to perform certain biological functions (i.e. kinase domain of kinases, or G-domain of signaling small GTPases). Phylogenies derived from protein domains can be safely used as proxies to approach protein phylogenies [Yang et al, 2005].

Signaling pathways appear to be essential in organisms requiring large regulatory functions to achieve particular physiological demands (from budding in bacteria, to cell cycle regulation checkpoints in eukaryotes). Essential pathways are phosphoryaltion and ubiquitination.

In this spirit we have analyzed  the presence of  potential phsophorylation modules, in particular the eukaryotic Serin/Threonin kinases (eSTKs) and  E3-ubiquitin ligases in prokaryotic genomes using a domain-based approach. To include species with highly regulatory demands, we selected members of the Planctomycetes-Verrucomicrobia-Chlamydia superphylum.

We have analyzed the domain content of several species, and compared their signaling domains distributions  in prokaryots showing special features (i.e.  inner membrane compartimentalization in Planctomycetes). We found a significant expansions of eSTKs in most of the members of the Phylum. Interestingly we found eukaryotic E3 members in few species, which suggests old horizontal gene transfers (Arcas et al, 2013). These findings aimed to instigate further investigations in the field of prokaryotes signalling.



Gene/proteins disease annotations

We also want to address integral aspects relevant to Personalized Medicine. With the novel technical developments and the exponential pace of data generation, based on large systematic screenings, a large list of “–omics” (genomics, epigenomics, transcriptomics, metabolomics, etc), it is our understanding that Integrative Biology is becoming a driving force to interrogate fundamental aspects in the Biomedicine field.  In this spirit, one of the objectives of IBIS is to understand the multidimensional aspects of disease mechanisms integrating large-scale and complex data. One of the key aspect of function prediction using graphs is gene/protein annotation. Since one of the main research lines involves disease and disease prediction, we have devoted our attention to create a high-quality, high-coverage, and low-redundancy database of gene-disease associations. This high confidence disease-associated genes act as “landmarks” in the functional map from which we can propagate function and predict new candidate genes. We have not only recorded the associations between genes and diseases, but also the molecular basis of the disease-causing mutation, and we have used control vocabularies and ontologies to annotate gene function, mutations and diseases. Although network-based approaches can be used globally to functionally annotate the whole genome, it is evident that signal propagation degrades with the distance to a trustable landmark. Thus, we have set to apply our methods, not for full genome annotation, but for most modest goal of prioritizing candidate gene lists. 

Other Projects: collaborations. 

Members of the group have been heavily involved in scientific collaborations at different levels. Using structure/sequence-based methods we have provided working hypothesis to guide experiments (Mañes et al, 2010) or structural interpretation of the impact of mutations (Athiyarath et al, 2013). The group has also developed web-servers to analyze iron-binding motifs (Campillos et al, 2010), as well as tailored tools to explore complex data (with Josep Manye at iGTP). 


Thursday the 23rd. . Powered by webtemplatesbox