gene tree output, what is the order of the treefile?

[CR Cardenas]

Hello,

I would like to use the 2.3k gene trees gene trees I've generated for some downstream analyses. For these I will need individual gene trees though. Yet, I don't know what loci in the partition those trees come from in the gene-tree.treefile

Here is the command I used to generate them:
iqtree -s trimalauto_30p.nex -S  nomerge_trimalauto_30p_parti.best_model.nex
-allnni -bb 1000 --prefix gene-tree

The resulting tree file has the expected. 2.3k gene trees. But, I have no clue what loci these trees came from in my partition file.

In the gene-tree.log there is a section where it says:
"SEPARATE TREE SEARCH FOR PARTITION"
Partition loc2026 / Iterations: 200 / LogL: -41305 / Time: 0h:17m:46s
Partition loci1001 / Iterations: 200 / LogL: -44744.3 / Time: 0h:18m:25s
Partition uce0900 / Iterations: 247 / LogL: -44235.4 / Time: 0h:22m:47s
Partition uce0666 / Iterations: 259 / LogL: -31930.8 / Time: 0h:23m:22s
....

I thought this might be the order of the trees in the output and hoped I could use them to create individual tree files. But the order between the partition ID, estimated model parameters, and tree search is inconsistent within the log file. I looked at the gene-tree.iqtree file which has another list of my partitions, and it does have the same partition ID information in the log and iqtree file.

To summarize my question, how do I tell what loci those gene trees come from when using -S and a partition file?

[CR Cardenas]

I'm just bumping this again, as I am genuinely curious. I ended up splitting up my partition file and building each tree individually with IQTree. But this seems excessive when there is a function for it in IQTree that does this already.

[CR Cardenas]

There is still no answer for this?

[Bui Quang Minh]

Hi Cardenas, The gene trees in treefile should be in the same order as the input partition file, which is also in the same order as written .iqtree file, which is also the same order in the beginning of log file, when IQ-TREE finishes reading the input files, before this line you wrote "SEPARATE TREE SEARCH FOR PARTITION" (This section may print in different order due to parallelisation scheduling, that tries to minimise the waiting time by doing tree search on largest partition first and smallest partition last).

Cheers,

Minh