HyPhy message board - dnds analysis with a rooted tree

	Welcome, Guest. Please Login

Home

Help

HyPhy message board › Methodology Questions › How to › dnds analysis with a rooted tree

(Moderators: Sergei, Simon)

‹ Previous Topic | Next Topic ›

Pages: 1

Send Topic

dnds analysis with a rooted tree (Read 2726 times)

line YaBB Newbies Offline I love YaBB 1G - SP1! Posts: 5	dnds analysis with a rooted tree Jul 17^th, 2007 at 6:18am Dear Sergei, I would want to know what could be the influence of tree topology on dndsbranch analysis. With the same data set I can have different results if I root the tree or not and I do not understand why... Thank you for you answer Line
Back to top	IP Logged

Sergei

YaBB Administrator

Offline

Datamonkeys are forever...

Posts: 1658
UCSD
Gender: male

Re: dnds analysis with a rooted tree
Reply #1 - Jul 19^th, 2007 at 4:12pm

Dear Line,

I am not certain which analysis you are referring to here.
Generally speaking, the rooting of the tree should have no effect on dN/dS estimates under time-reversible models, so what you seeing may be a bug.
I'd like to ask you to help me recreate the situation, so that I can give you a meaningful answer...

Cheers,
Sergei

Associate Professor
Division of Infectious Diseases
Division of Biomedical Informatics
School of Medicine
University of California San Diego

WWW

IP Logged

line

YaBB Newbies

Offline

I love YaBB 1G - SP1!

Posts: 5

Re: dnds analysis with a rooted tree
Reply #2 - Jul 23^rd, 2007 at 9:21am

dear Sergei,
I use the testbranchDNDS.bf, and the results are different. In fact I do not really no what kind of test I have to use. My gene phylogeny put in evidence two clades, one present long branch length, it is organised by species and it presents good bootstrap scores whereas the other is the contrary. In fact I would like to know if these different evolution is due to different selective pressure. I do not know if it is better to use DNDSrateanalysis or testbranchDNDS.bf, or the two. For me the first test is more scientific because we make no hypothesis (expect for the model etc...) whereas in the second we have to specify branch supposed to evolve differently. Is the testdNdSbranch.bf could be a test to be sure that gene lineages evolve under different selective presssure?
I am not really sure that I am clear, but I hope you can understand!
Thank you for your help
Line

IP Logged

Sergei

YaBB Administrator

Offline

Datamonkeys are forever...

Posts: 1658
UCSD
Gender: male

Re: dnds analysis with a rooted tree
Reply #3 - Jul 27^th, 2007 at 11:54am

Dear Line,

If you want to compare average strengths of selection between two clades separated by a single branch (known), try running the SelectionLRT.bf analysis in Compartmentalization (under standard analyses). We used this test in a viral evolution setting very similar to what you are describing (Multimedia File Viewing and Clickable Links are available for Registered Members only!! You need to

Login see figure 1 and discussion around it).

If you wish to scan your tree for evidence of variable selection without a hypothesis in hand, give Multimedia File Viewing and Clickable Links are available for Registered Members only!! You need to

Login a try.

Also, take a look at Multimedia File Viewing and Clickable Links are available for Registered Members only!! You need to

Associate Professor
Division of Infectious Diseases
Division of Biomedical Informatics
School of Medicine
University of California San Diego

WWW

IP Logged

Pages: 1

Send Topic

‹ Previous Topic | Next Topic ›

« Home

‹ Board

Top of this page