"Dr. Christoph Gille" <[email protected]> writes: > I want to verify or reject the hypothesis that there is a yet unknown > putative reading frame in a known coding viral gene.
One caveat: I haven't worked with viral sequences. > The translated amino acid sequence does not yield any blast or prosite > hits. > Question: how can I verify by computational methods, that this could > be indeed an additional coding reading frame, resulting in an amino > acid sequence for which no similar sequence exist in todays databases. > I would first check in related sequences that the reading frame is > indeed open. What related sequences do you have? If you have other strains of the virus, conservation of the amino sequence (Ka/Ks ratio) might indicate a translated sequence. > Then I could look for irregularities in codon usage resulting from the > two overlapping reading frames. Is there enough codon bias that this gives useful evidence? > Then I might observe absense of base wobbling in the third position > due to the other reading frame. > Can you recommend any search for functional sites? > All methods based on sequence similarity will fail since there > is no similar sequence so far. I think that if you have conserved functional sites, they would show up as short BLAST hits as well. Perhaps HMM-based methods can be more sensitive. I don't think any purely computational method is going to be a solid indicator for this, is it an option to try to capture the putative transcript with PCR? Or use some other in vitro method? -k -- If I haven't seen further, it is by standing in the footprints of giants _______________________________________________ BBB mailing list [email protected] http://www.bioinformatics.org/mailman/listinfo/bbb
