Comments on: HMMER3 testing notes – my skills are (finally) becoming obsolete

By: Pawel Szczesny

Pawel Szczesny — Sat, 02 May 2009 08:49:00 +0000

Thank you for the interesting comment dr Bateman, I’ve indeed missed the publication about SIMPRO.

As for my sequence comparison benchmark, I’ll prepare a separate post about that, but in short: it’s not a large scale benchmark – more of a CASP-like (I have access to a few structures that are not yet published). Also, I don’t mind false positives (even in cases I don’t know the answer I usually can filter these out by looking at the alignment), so it’s very likely that I sort tools in different order that most of other people.

By: alexbateman

alexbateman — Thu, 30 Apr 2009 09:48:55 +0000

Thanks for the interesting article. I’d be interested to read more about your sequence comparison benchmark. I found that PRC outperformed HHsearch on my Pfam Clans benchmark [1]. You might also be interested to know about a recent publication that appeared to significantly improve the performance of HHsearch by looking at a network of matches [2].

References

1. Bateman A, Finn RD. SCOOP: a simple method for identification of novel protein superfamily relationships. Bioinformatics. 2007;23(7):809-814.
2. Jung I, Kim D. SIMPRO: simple protein homology detection method by using indirect signals. Bioinformatics. 2009;25(6):729-735.

By: Richard Karpinski

Richard Karpinski — Wed, 22 Apr 2009 14:24:35 +0000

I was fascinated to discover the link from several unique forms of an apparently unrelated gene in a few families to subsequent muscular distrophy. The middle of the RNA became “sticky” and hybridized with RNA of a signaling protein affecting expression of a hundred or so other genes which are related to the distrophy. Accidental creation of a double helix of RNA acting as RNAi. Perfectly sensible once you think of it, but weird.