Computational inference of functional relationships from multiple genomes : Drosophila carbonic anhydrases.

Abstract

Carbonic anhydrases (CAHs) are metalloenzymes which catalyze the reversible chemical reaction between carbon dioxide and water to bicarbonate and a proton, which is crucial in pH homeostasis. There are several CAH families in different protein classes with independent evolutionary origins. In vertebrates, at least 15 isoforms exist, classified into cytoplasmic, mitochondrial, secreted, and membrane-associated CAHs, and acatalytic CAH-related proteins. Invertebrates have a similar multiplicity in their alpha CAH genes, but most of the isoforms have no direct equivalents within vertebrate CAHs. Vertebrate CAHs are well characterized for their tissue distribution and functions, but no functional analyses of a complete set of all CAH isoforms have been made in any invertebrate species. This study deals with in silico functional prediction based on the evolution of 15 alpha CAH ortholog groups in 22 complete drosophila genomes. Additionally, molecular evolution analysis detected no significant positively selected sites in the sequence of invertebrates. Different bioinformatics analyses were done to predict the subcellular localization, multiple sequence analysis, phylogenetic analysis, and alpha CAH were functionally categorized which was shown to be reliable to the expression data from FlyBase. The gene set whose gene expression is specific for male reproductive organs, for the spermatheca, have found to be highly statistically correlated. The molecular evolution analysis of CAH gene family in invertebrate detected no significant positively selected sites by BEB analysis in PAML. This novel bioinformatics method for functional grouping has similarity with the expression data of drosophila species. This method can be further modified and used to identify the functional classification of other gene families.