MODEL TREES

Moisès Gil i Bernabeu

unread,

Dec 11, 2019, 1:25:24 PM12/11/19

to IQ-TREE

Hi! I want to compare topologies of several models for a class. I would like to know if there is any possibility of writing trees of different models computed by IQ-Tree in an output file. I'm currently using both 1.6.12 and 2.0 versions.

Thank you!!

Minh Bui

unread,

Dec 11, 2019, 6:04:24 PM12/11/19

to IQ-TREE

ModelFinder fixes the tree topology for model selection. So for all models evaluated, the tree is the same (while the branch lengths are different).

Minh

On 12 Dec 2019, at 5:25 am, Moisès Gil i Bernabeu <moigil....@gmail.com> wrote:

Hi! I want to compare topologies of several models for a class. I would like to know if there is any possibility of writing trees of different models computed by IQ-Tree in an output file. I'm currently using both 1.6.12 and 2.0 versions.

Thank you!!

--
You received this message because you are subscribed to the Google Groups "IQ-TREE" group.
To unsubscribe from this group and stop receiving emails from it, send an email to iqtree+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/iqtree/08b50190-5996-4e77-8884-7001ed23a5ad%40googlegroups.com.

Moisès Gil i Bernabeu

unread,

Dec 12, 2019, 4:02:26 AM12/12/19

to IQ-TREE

Yes, but, is there any output file with these lengths? I've seen .treels have a lot of information, but I can't relate it with concrete model. Tanks for the answer!

El dijous, 12 desembre de 2019 0:04:24 UTC+1, Minh Bui va escriure:

ModelFinder fixes the tree topology for model selection. So for all models evaluated, the tree is the same (while the branch lengths are different).

Minh

On 12 Dec 2019, at 5:25 am, Moisès Gil i Bernabeu <moigil....@gmail.com> wrote:

Hi! I want to compare topologies of several models for a class. I would like to know if there is any possibility of writing trees of different models computed by IQ-Tree in an output file. I'm currently using both 1.6.12 and 2.0 versions.

Thank you!!

--
You received this message because you are subscribed to the Google Groups "IQ-TREE" group.

To unsubscribe from this group and stop receiving emails from it, send an email to iqt...@googlegroups.com.

Minh Bui

unread,

Dec 13, 2019, 7:16:37 AM12/13/19

to IQ-TREE, Moisès Gil i Bernabeu

Hi there,

There is no such option…so I’ll need to run IQ-TREE for each model manually. The treels file (printed by -wt option) contain the trees encountered during tree search. So it has nothing to do with the model selection.

Cheers,

Minh

To unsubscribe from this group and stop receiving emails from it, send an email to iqtree+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/iqtree/173bcb0a-2f9f-4ab8-afe2-812ecd4f1440%40googlegroups.com.

Brian Foley

unread,

Dec 14, 2019, 11:03:59 AM12/14/19

to IQ-TREE

The TreeRate tool at the HIV Databases:

https://www.hiv.lanl.gov/content/sequence/TREERATEv2/treerate.html

allows you to past in a Newick format treefile (such as output from IQtree) and then it creates a table of branch lengths.

It gives you distance from a not to every other node and to each of the tips.

So you could save each table to distances into a spreadsheet or file and then run some computations to compare them (sum total of distances; ratio of tip distances to intra-node distances, etc...

Brian Foley

unread,

Dec 14, 2019, 11:10:23 AM12/14/19

to IQ-TREE

The HIV Databases also has an IQtree version so you can run IQtree there and at the bottom of the results page you select BranchLength to analyze the branch lengths. I also recommend saving the log files from IQtree to look at other parameters such as the rate matrix.

Reply all

Reply to author

Forward