Computational Genomics

The Program of Computational Genomics is focused, in the field of bioinformatics, on the challenge of accelerating access to genomic knowledge, in particular on the regulation of microbial gene expression. The backbone of our work is the design and expansion of the RegulonDB database, with knowledge of the genetic regulation in Escherichia coli K-12 extracted by curators, both from classical experiments and from High-Throughput (HT) post-genomic methods. We work with algorithms for the identification of binding sites, promoters and transcription units, as well as co-regulation, and for the processing of data from HT methods, such as ChIP-seq, among others. We work in an ontology and a controlled vocabulary of microbial growth conditions, and in an ontology of microbial genetic regulation. An important aspect is the use of Natural language processing (NLP) to enrich knowledge extraction strategies that shall allow us to expand the biocuration to other bacteria such as Salmonella, and other few more bacteria. We have supported integrative concepts such as regulatory phrases and, more recently, the genetic-metabolic sensory-response units (GENSOR Units), which pose a lot of new questions. We are investigating the formalism of S-systems by Michael Savageau for the modeling, initially, of GENSOR Units; extending it to complete circuits and, if possible, to an integral modeling of the cell at the molecular level. This with a novel approach of computational algebraic geometry.
All this way of seeing and conceiving scientific knowledge has its application with similar strategies in a project that seeks to break the separation between texts for laymen and texts for experts with the Conogasi project (conogasi.org).

Group Staff

Group Publications