Clustering of protein domains for functional and evolutionary studies

An evolutionary split statistic is calculated for each column in a protein multiple sequence alignment; the statistic has a larger value when a column is better described by an evolutionary model that assumes clustering around two or more amino acids rather than a single amino acid. The user selects columns (typically the top ranked columns) to construct a motif. The motif is used to divide the family into subtypes using a stochastic optimization procedure related to the deterministic annealing EM algorithm (DAEM), which yields a specificity score showing how well each family member is assigned to a subtype. The clustering obtained is not strongly dependent on the number of amino acids chosen for the motif. The robustness of this method was demonstrated using six well characterized protein families: nucleotidyl cyclase, protein kinase, dehydrogenase, two polyketide synthase domains and small heat shock proteins. Phylogenetic trees did not allow accurate clustering for three of the six families.The method clustered the families into functional subtypes with an accuracy of 90 to 100%. False assignments usually had a low specificity score.Rapid progress in DNA sequencing is generating large numbers of deduced protein sequences. The prediction of their function is an important problem in Bioinformatics. This is tackled by comparing new sequences to known sequences as high sequence similarity usually indicates related function. It is possible to use similarity search algorithms such as BLAST [1]. A more sensitive approach is to use hidden Markov models (HMMs) to define protein families as implemented in HMMER suite of programs [2]. Such HMM profiles are used to define protein families in the Pfam database [3]. In many cases, these families consist of functional domains in larger proteins.In many cases protein families can be split into sub-types based on functional differences e.g. substrate specificity such as for the malonyl-CoA- and methylmalonyl-CoA-incorporating acy