IQ-TREE 1.6.8: Data type detection problems (AA meta-partitions)

13 views
Skip to first unread message

Jan Philip Oeyen

unread,
Jun 19, 2019, 10:04:02 AM6/19/19
to IQ-TREE
Hi,
I am analyzing a small AA alignment (15 species, ~330 meta-partitions). When running IQTREE 1.6.8 (newest version currently available on our cluster), the software automatically detects the type accurately for about 2/3 of the partitions if I do not use the -st AA option (see copy&paste from log-file below). I ran the analysis both with and without the data type option and got the same topology with overlapping likelihoods and only very minute variations in branch lengths (20 runs without and 1 run with -st AA). What does IQTREE do with the partitions for which it was not able to automatically determine the datatype when not using the -st option? Is this something one should look out for more carefully? I searched through the documentation, but could not seem to find this specific case.

Best,
Jan Philip Oeyen

Subset  Type    Seqs    Sites   Infor   Invar   Model   Name
1               15      920     385     279     LG4X    EOG090X0LFT_EOG090X08KD_EOG090X0EXU_EOG090X0KP6
2       AA      15      1903    135     1616    LG+R2   EOG090X0GN3_EOG090X0064_EOG090X07IC
3               8       238     72      75      LG4M    EOG090X0C5G
4       AA      15      305     141     71      LG+R3   EOG090X0EYV_EOG090X0EV6
5       AA      15      887     244     386     LG+F+R3 EOG090X0ACT_EOG090X0K4K_EOG090X03QT
6       AA      15      584     34      458     JTT+R2  EOG090X0662_EOG090X09VR_EOG090X0KYU

Minh Bui

unread,
Jun 25, 2019, 11:23:41 AM6/25/19
to IQ-TREE, Jan Philip Oeyen
Hi Jan,

This is part of the log file, where the detection might not print complete information. You should look at .iqtree file with full information, e.g.:

SEQUENCE ALIGNMENT
------------------

Input data: 17 taxa with 3 partitions and 1332 total sites (0% missing data)

  ID  Name   Type       Seq     Site    Unique  Infor   Invar   Const
   1  part1  DNA        17      666     413     343     227     227
   2  part2  DNA        17      333     208     153     112     112
   3  part3  DNA        17      333     207     165     127     127

And simply look at “Type” column, to see if these are all correct or not.

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.
To view this discussion on the web visit https://groups.google.com/d/msgid/iqtree/a7e59ca2-b19f-4c87-950d-e71084e730d0%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply all
Reply to author
Forward
0 new messages