Generate MrBayes block from IQ-TREE

[Dave Id]

Is there some way to generate lset and prset commands in a MrBayes data block for each partition that was generated using IQ-TREE's built in PartionFinder3? I'm basically looking for something that has the functionality of PhyloSuite or partitNex2MrBayes, both of which are still running PartitionFinder2 and are very slow off the cluster.

Specifically, I used these commands in IQ-TREE:
iqtree2 -nt 6 -bnni -s infile.txt -p partition_file.txt -st DNA -mset mrbayes -merit AICC -m MFP --sprrad 6 -B 1000 

This was the output in the best_scheme.nex file:#nexus
begin sets;
  charset ITS = 1-795;
  charset LSU = 796-1753;
  charset RBP1-1 = 1754-1946;
  charset R1Intron1 = 1947-2075;
  charset RBP1-2 = 2076-2823;
  charset R1Intron2 = 2824-2886;
  charset RBP1-3 = 2887-3503;
  charset RPB2-1 = 3504-3596;
  charset RIntron1 = 3597-3661;
  charset RPB2-2 = 3662-4171;
  charset RIntron2 = 4172-4248;
  charset RPB2-3 = 4249-5247;
  charset Tef-1 = 5248-5451;
  charset TIntron1 = 5452-5518;
  charset Tef-2 = 5519-5655;
  charset TIntron2 = 5656-5721;
  charset Tef-3 = 5722-5864;
  charpartition mymodels =
    GTR+F+I+G4: ITS,
    GTR+F+I+G4: LSU,
    SYM+I+R2: RBP1-1,
    GTR+F+G4: R1Intron1,
    GTR+F+I+G4: RBP1-2,
    K2P+I: R1Intron2,
    SYM+I+G4: RBP1-3,
    K2P+I+G4: RPB2-1,
    GTR+F+R2: RIntron1,
    SYM+I+G4: RPB2-2,
    GTR+F+R2: RIntron2,
    SYM+I+R3: RPB2-3,
    GTR+F+I+R3: Tef-1,
    HKY+F+I+G4: TIntron1,
    GTR+F+I+G4: Tef-2,
    HKY+F+I+G4: TIntron2,
    GTR+F+I+G4: Tef-3;
end;

Thanks in advance!

[Denis Zabin]

Following, I'd also love if there was a possibility of IQTree automatically generating the Nexus file with the model parameters for each partition. As of now, I use the .ckp file generated to include the model parameters in the nexus file.

[Mark Miller]

Hi all,

Thanks for the interest in this issue.

I believe it is probably most efficient to direct this request to the IQTree developer group, https://groups.google.com/g/iqtree

It seems to me that they may be able to modify the output to meet your needs.

I will look into what can be done on our side vis a vis the lset prset issue, but it will certainly require manual editing of the MB block, which seems tedious and error prone at this stage.

If you have a specific failure message from MrBayes on CIPRES, just send me the file_JOBINFO.TXT from one or more such jobs, and I will investigate.

If the IQ TREE group responds, I will follow up with their recommendations.

Best,
Mark

[Mark Miller]

Sorry, IQ discussions has moved to github

https://github.com/iqtree/iqtree2/discussions

Mark

[Denis Zabin]

Thank you, Mark. Looking forward to any answer on this topic from the devs.

[Ziv Lieberman]

Hi Dave, Denis,

I am not very familiar with MrBayes so I'm not sure if this helps answer your questions, but the IQ-TREE output file ending in best_model.next has the model parameters included for each partition, unlike best_scheme.nex which only has the selected models, e.g.

. . .

charset uce_1395_left.nexus_uce_734_right.nexus = 1075644-1075726  1353055-1353268;
  charpartition mymodels =
    TPM2{1.22621,7.1967}+F{0.266978,0.23961,0.271723,0.221689}+I{0.139347}+G4{0.956881}: uce_10001_core.nexus_uce_10079. . .

Best,

Ziv

--
You received this message because you are subscribed to the Google Groups "CIPRES Science Gateway Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cipres-science-gatew...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cipres-science-gateway-users/860af53a-7b83-44bf-b317-ff8402869f4fn%40googlegroups.com.

[Dave Id]

Thanks Mark, I opened a discussion here: https://github.com/iqtree/iqtree2/discussions/191

[Mark Miller]

Great, thanks for doing that.

As I understand it currently, we have to understand and specify the lset and prset parameters for each of the models within the mb block.

Unfortunately, I am more on the tech side, and less on the domain side of this, so I cant do that off the top of my head. 

However, if you can affirm that you believe this is correct, I can try to educate myself on those paramters.

Of course if its wrong that would also be helful to know... :)

Mark

[Dave Id]

Hi Mark,

Yes I believe that is correct. There are tables of MrBayes model parameters and corresponding lset and prset commands. I'm not 100% sure where to find the most reliable table though.

I since raised a feature request, which just needs a handful of people support to become implemented.
Thanks, David

[Dave Id]

Here is the link: https://github.com/iqtree/iqtree2/issues/195

[Miller, Mark]

I can read the issue, and gave ti thumbs up, and subscribed to it.

Thanks for detailing the issue so elegantly.

 

Mark

--

You received this message because you are subscribed to the Google Groups "CIPRES Science Gateway Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cipres-science-gatew...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/cipres-science-gateway-users/bc12efcf-b2e7-4074-aaf9-92ed012e5a6fn%40googlegroups.com.