Open njdowdy opened 5 months ago
Hey there, thanks for giving pipesnake
a run.
In retrospect, choosing four samples of the same species for the example data was probably not my cleverest move. I'll try and swap those example data out so that this issue doesn't happen to folks in the future, thanks for bringing it to our attention. Will close this once I've replaced those samples.
Cheers.
I think it would be useful to make the PipeSnake workflow log and then throw out any loci that are invariant for all taxa, even if it is in the set of targets. I can imagine cases where we may not know a priori whether all the targets are going to be well-behaved or not.
I'm looking forward to digging into PipeSnake more!
Description of the bug
Running the test dataset on initial install yields the IQTREE error:
This is expected behavior for IQTREE (see issue: https://github.com/Cibiv/IQ-TREE/issues/155), however this causes PipeSnake to exit, failing to complete the remainder of the pipeline.
I fixed this by appending the
-keep-ident
option to the IQTREE command on Line 38 inmodules/local/iqtree.nf
(see IQTREE General Options for more info on-keep-ident
). However, it would be better if PipeSnake handled the issue in a better manner, as using-keep-ident
may increase the run time of the analysis per Heiko Schmidt here and I think making the user explicitly aware of this issue would be better.Command used and terminal output
Relevant files
No response
System information
Nextflow version 23.10.1 build 5891 Hardware: HPC Executor: local Container engine: Conda in Docker OS: Debian 12 Version of ausarg/pipesnake 1.1