The MIp Toolset: an efficient algorithm for calculating Mutual Information in protein alignments

Abstract

Background: Coevolution within a protein family is often predicted using statistics that measure the degree of covariation between positions in the protein sequence. Mutual Information is a measure of dependence between two random variables that has been used extensively to predict intra-protein coevolution. Results: Here we provide an algorithm for the efficient calculation of Mutual Information within a protein family. The algorithm uses linked lists which are directly accessed by a pointer array. The linked list allows efficient storage of sparse count data caused by protein conservation. The direct access array of pointers prevents the linked list from being traversed each time it is modified. Conclusions: This algorithm is implemented in the software MIpToolset, but could also be easily implemented in other Mutual Information based standalone software or web servers. The current implementation in the MIpToolset has been critical in large-scale protein family analysis and real-time coevolution calculations during alignment editing and curation. The MIpToolset is available at: https://sourceforge.net/projects/miptoolset/