Two articles in last year’s Bioinformatics popped up on MyNCBI alerts this week that are worth mentioning: both deal with evolutionary aspects of protein structure.
The first, How old is your fold?, by Winstanley et al., attempts to reconstruct the phylogenetic relationships and ages of conserved protein folds in major domain superfamilies. Their method constructed parsimonious occurrence patterns of folds in a broad phylogenetic representation of species, and correlated to previously used protein age estimators. Winstanley et al. note that problems are few, but caused mainly by (1) lack of tree definition in the early bacteria and (2) variation of parsimonious scenario assignment between origin in the Last Universal Common Ancestor (LUCA) and origin at the top superkingdom node.
So, Winstanley et al. isn’t groundbreaking in any major way, but it’s a novel and interesting approach, to be sure.
The second paper, Predicting protein-protein interaction by searching evolutionary tree automorphism space, by Jothi et al., introduces MORPH, a new algorithm for predicting protein interaction partners betwen members of two protein families that are known to interact. The main benefit of this co-evolution-based method is to exploit the tendency for interacting proteins to have similar phylogenetic trees. While, I confess, I’m not too familiar with similarity matrices, or bioinformatic algorithms in general, the notion that the co-evolution of proteins can be elucidated is almost a why didn’t someone think of this sooner? situation.
Also, Jothi et al. appears quite complimentary to Bridgham et al.. Neat.
- How old is your fold? Winstanley HF, Abeln S, Deane CM. Bioinformatics. 2005 Jun;21 Suppl 1:i449-58. Pubmed.
- Predicting protein-protein interaction by searching evolutionary tree automorphism space. Jothi R, Kann MG, Przytycka TM. Bioinformatics. 2005 Jun;21 Suppl 1:i241-50. Pubmed.
- Evolution of hormone-receptor complexity by molecular exploitation. Bridgham JT, Carroll SM, Thornton JW. Science. 2006 Apr 7; 312(5770):97-101. Pubmed.