Reducing Space and Time

I recently read “The Treeterbi and Parallel Treeterbi algorithms: efficient, optimal decoding for ordinary, generalized, and pair HMMs” by Keibler, Arumugam and Brent which was published in Bioinformatics 23(5):545-554 (2007). This paper outlines an algorithm for optimal decoding of HMMs which reduces memory requirements compared to the more standard Viterbi decoding. Furthermore, unlike the Hirschberg algorithm (aka Myers-Miller) it does not result in a significant increase in runtime.

News: Link fest.

I’ve been mentioned on Bio::Blogs #17 (the one Santa brought us)! This has provided the necessary motivation to mention some good posts I’ve read recently on Blogs:

Pedro Beltrao over at Public Rambling has a good post on the recent Cell paper by Lécuyer on Global Analysis of mRNA Localization. This paper has been on my radar because I’ve been working with the BioGRID lately and therefore thinking a bit about how to infer missing network connections.

The recent article at PLoS ONE by Keller and Harel entitled “Beyond the Gene” attracted some attention in the blogs I read. Jason Stajich over at Fungal Genomes and Comparative Genomics highlights some problems with the jargon introduced by Keller (post here) and RPM at evogen rips the article to shreds on a number of grounds, including the dreaded “adaptionist claptrap” (post here). I can honestly say that I got more from reading these two posts than from the article itself. (I reviewed a good article related to the “adaptionist claptrap” in my very first post.)

Finally, while not a literature review, I enjoyed the recent post Good Programming versus Biological Intuition over on Bioinformatics Zen.