Good morgon!
I have a set of HBV mutants in a region where overlapping occurs between two genes (p2/s1, the s and the polymerase gene, which evolve independently as stated by Zaaijer et al., J Gen Virol 2007, 88, 2137). The SBP analysis gave me a possible breakpoint, but the complete analysis for recombinant was impossible (output: cAIC could not be applied to this alignment; it has too few sites for the number of sequences). So. I performed the analysis with only one tree and had quite good results in comparison with tree obtained by bootstrapped ML.
My questions are:
1. could the breakpoint reflect something due to overlapping?
2. How can i introduce in my analysis a correct restrain for the overlapping?
3. Evaluating the antigenic escape is the most important aim of this work: I have 107 sequences, exceeding the capacity of TOGGLE. Could I submitt to TOOGLE only 100 selected sequences without bias (e.g. eliminating the WT seq and those with dN/dS=1)? in this case can I use TOOGLE independently from the other classical analyses offered by Datamonkey (i.e. SLAC and FEL)?
Thank you in advance, especially for data i have already get.
Best wishes.  ::)
Anna Giulia

Hi Anna,

I haven't read the manuscript you cite, but based on my experience with the analysis of dual coding genes in mammals (see Multimedia File Viewing and Clickable Links are available for Registered Members only!!  You need to Login) I find it hard to imagine that the reading frames are evolving independently. To answer your questions:

1). I would expect SBP/GARD to pick up breakpoints at the boundary of the dual coding region -- are you seeing one in the middle of it? You might see a significantly suppressed evolutionary rate in the dual coding region compared to the flanking regions.
2). The only model in HyPhy is described in the PLoS comp bio paper I cite above, and even that is a bit of a hack. The model in the paper is very memory hungry -- so it may take a long time to run on 100+ sequences.
3). I've increased the limit to 125 sequences, so you should be able to run your entire data set through TOGGLE.

Sergei

Hi, Sergei,
I apologize I did not answer earlier: after receiving your suggestions I was unable to use the server, for one of those strange tinks happening in the net - sometimes.
Today I have finally done Toggle: thank you very much for having improved the capacity of your software!
Now I have planned to be brave enough and try to put the double code of my overlapping genes in some software, e.g. in HyPhy. If I have troubles, I will return again to ask your help, Thanks!
(I am looking in the middle of the two genes).
:D
Best wishes.
Anna Giulia