BLINDTest is a basic research project involving three collaborative partners: two research teams and a bioinformatics platform. Through a comparative transcriptomics approach, we would like to identify mutations involved in the physiological, morphological and behavioral adaptive evolution of cavefish populations of the species Astyanax mexicanus. Indeed, the long term occupancy by surface river-dwelling fish of an open and empty ecological niche such as the cave system implies two major adaptive challenges: finding food and finding mates in the total and permanent darkness. The main reason why we choose to study this species is that it presents two very different types of morphs. The first corresponds to normal, river-dwelling fish of Central and South America. The second corresponds to a few dozens of populations (29 known today) of blind and de-pigmented fish which inhabit caves of the Sierra de El Abra region in Mexico. Cave colonization occurred several times and in an independent manner, since about 1 million years. Some cavefish population may be relatively recent, while others are more ancient, but they all correspond to the parallel evolution of several phenotypic traits in relation to the life in caves. In this project, we propose to compare the transcriptomes of surface fish and three independently-evolved cavefish populations and analyze them, in order to answer the three following questions: 1) What is the extent and what is the nature of structural variations in the transcriptomes of cave-adapted versus surface populations? 2) Can we find some examples of parallel evolution at transcriptome level in independently-evolved natural populations? 3) Are the structural transcriptome changes responsible for the physiological, morphological and behavioral adaptation to caves observed in cave populations? BLINDTest will take advantage of the possibilities offered by second generation sequencing technologies to describe transcriptome-wide the genetic polymorphism that arose within and between natural populations of a vertebrate species presenting populations living in radically different environments. We will then further test functionally the adaptive nature of observed genetic changes through a functional approach including crosses between populations of fish, transgenesis experiments, as well as developmental and behavioral analyses. This work, which will also include the identification of thousands of selectively neutral SNPs, will constitute a solid basis for future population genetics studies and for testing hypotheses of selection at a few loci (for instance selective sweep) and searching QTLs.