HyPhy message board - breaking at breakpoint?

	Welcome, Guest. Please Login

Home

Help

HyPhy message board › Methodology Questions › How to › breaking at breakpoint?

(Moderators: Sergei, Simon)

‹ Previous Topic | Next Topic ›

Pages: 1

Send Topic

breaking at breakpoint? (Read 1507 times)

sunilkalmadi

YaBB Newbies

Offline

Feed your monkey!

Posts: 15

breaking at breakpoint?
Apr 6^th, 2008 at 6:17pm

Hi,
I have 3 different questions. Kindly answer them.

1. Whether it is valid to cut the sequence dataset into pieces at the breakpoints obtained from singlebreakpoint.bf(GARD) and perform hyphy standard analyses (FEL, REL, subtree etc) independently on these pieces? I am interested because as of now GARD input are only considered by branchclassDNDS.bf.

2. Site wise selection methods like REL and FEL on datamonkey donot have option for foreground branch to be selected. I am interested to know whether the first sequence or the whole dataset itself is considered as the foreground branches in these analyses

3. This is rather a theorectical question. I would like to know whether effect of different branch lengths on the codon model can be analyzed using AIC values, provided same dataset and tree topology.

Thanking you
Regards,
Sunil.

IP Logged

Art Poon

Global Moderator

Offline

Feed your monkey!

Posts: 0

Re: breaking at breakpoint?
Reply #1 - Apr 7^th, 2008 at 12:41pm

Hi Sunil,

1. Performing FEL, REL and other analyses on alignment partitions by breakpoints identified with a single breakpoint analysis should be quite valid, so long as there is sufficient variation within the partition.

2. REL and FEL, by default, do not model branch-specific variation in substitution rates; all branches are assumed to evolve at the same rates, all else being equal (branch lengths, site-specific variation).

3. Having a 'bad' set of branch length estimates, given the same data set and topology, would correspond to a poorly-optimized model. AIC would only become relevant if we were attempting to distinguish between alternative models, e.g., if we introduce rate variation over time (such as a covarion model) in an alternative model.

Cheers,
- Art.

IP Logged

sunilkalmadi YaBB Newbies Offline Feed your monkey! Posts: 15	Re: breaking at breakpoint? Reply #2 - Apr 7^th, 2008 at 2:10pm Dear Artpoon, Thanks a lot for your answers. Regards, Sunil
Back to top	IP Logged

Pages: 1

Send Topic

‹ Previous Topic | Next Topic ›

« Home

‹ Board

Top of this page