On 1/19/10 1:34 PM, Ben Temperton wrote:
I am currently writing a pipeline to identify distant homologs of genes in metagenomic samples using HMMER 3.0 and hmm models of conserved domains from Pfam. Once I have identified putative homologs, I would like to verify that these homologs are indeed homologs by threading them onto a known protein structure and evaluating whether the threading was successful. Given I'm very new to protein threading, is Modeller an ideal tool to use for this application?
You could probably use it, but I don't know if it would be the best tool for the job. Given an alignment between a known structure and a sequence, Modeller will always build a model. You could certainly use an assessment score such as normalized DOPE to estimate whether the model is a good one, but I'm not sure where you'd pick your cutoff - it would be something for you to tinker with.
Ben Webb, Modeller Caretaker
--
http://www.salilab.org/modeller/
Modeller mail list: http://salilab.org/mailman/listinfo/modeller_usage