“ERROR: Rate entries could not be read” when using custom codon exchange rates and frequencies

[Umberto Perron]

Hi, I'm runnin iqtree v1.5.5. for Linux 64-bit ad issuing the following command:

$ iqtree -s testalign -st CODON -m testmodel

where "testmodel" is a file containing custom codon exchange rates and frequencies in PAML format.
Unfortunately it results in the aforementioned error.
I have also tried to supply the ECM matrix as presented in the supplementary materials of Kosiol,C., Holmes,I., Goldman,N., 2007 (doi) whith identical results.
I'm quite sure that I'm missing something basic but I have no idea how to solve this.
Best,

Umberto

[Bui Quang Minh]

Dear Umberto,

The documentation (http://www.iqtree.org/doc/Substitution-Models#codon-models) about user-defined codon model is wrong. For standard genetic code the file should contain 61*61 symmetric rate matrix and 61 frequencies (not in PAML format). I will update the doc, thanks!

Note that the codon order is AAA,AAC,AAG,AAT,ACA,….,TTT, excluding the 3 stop codons. I believe that PAML uses a different order. So pls check this carefully.

Moreover, your codon model is problematic: it has many codons with zero frequency. This will cause numerical problem. So please change this, before you run IQ-TREE. I don’t know how small a frequency can get before making problem, but it should not be smaller than 0.0001.

Cheers

Minh

--
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 post to this group, send email to iqt...@googlegroups.com.
Visit this group at https://groups.google.com/group/iqtree.
For more options, visit https://groups.google.com/d/optout.
<testmodel><testalign.log>

--

Bui Quang Minh

Center for Integrative Bioinformatics Vienna (CIBIV)

Campus Vienna Biocenter 5, VBC5, Ebene 1

A-1030 Vienna, Austria

Phone: ++43 1 4277 74326

Email: minh.bui (AT) univie.ac.at