Research Lines in Infogenomics

Infogenomics intends to develop genome analyses based on a pure informational perspective. The main objectives of Infogenomics[1,2] are the following: i) the definition of parameters, with their computation in real genomes, that can discriminates interesting genomic features and properties, ii) the definition, with their computation in real genomes, of statistical distributions providing interesting profiles associated to genomic elements, and finally iii) the extraction of informationally relevant genomic words, in genomes or classes of genomes. Concerning the last point, many possible methods can be developed, and an adequate notion of informational relevance needs to be elaborated. Some ingredients are important to this end. Firstly, the basic notions of Shannon information theory (and their following elaborations) as information source, entropy, entropic divergences, mutual information, randomness and typical distribution of random processes (Gaussiam, Poisson, and exponential distributions). Moreover, basic concepts from formal language theory are useful, in order to apply probabilistic concepts to strings and sets of strings. “Short” strings are the main constituents of the “long” strings constituting genomes, and surely some of them are the components of genome constructions and developments, during the evolutive process that produced them (assembling longer sequences from some initial minimal genomes, or RNA genomes of proto-organisms