Search failures with searchcli docker

[fiuzatayna]

Greetings,

I'm trying to run searchcli with several software but unfortunately only got successfull messages from 4/11 of them.

I'm using quay.io/biocontainers/searchgui:4.3.0--hdfd78af_0 searchgui eu.isas.searchgui.cmd.SearchCLI

Successful runs

👍myrimatch
👍msgf
👍comet
👍xtandem

Failed

❌ tide

Tide

The main issue regarding is in the line below. I could try to add on top of searchgui's docker image and add this library, but I just want to make sure it can't be fixed by another much easier change.

/usr/local/share/searchgui-4.3.0-0/resources/Tide/linux/crux: error while loading shared libraries: libstdc++.so.6: cannot open s
                                                                                                                                shared object file: No such file or directory

No information on execution / failure / success

❔meta_morpheus
❔sage
❔novor
❔directag
❔ms_amanda
❔omssa

When I run those last by themselves the search finishes too fast.
Then, as expected, when I try to run PeptideShaker I'm informed there are no identifications.

[fiuzatayna]

I updated the text to reflect behavior with quay.io/biocontainers/searchgui:4.3.0--hdfd78af_0, as version 4.2.17 returned failed execution for ms_amanda and omssa as well.

[hbarsnes]

I'm afraid that we do not work with Docker containers ourselves and therefore do not have much experience with them. My advice will therefore be of a more general type.

The first thing I would do is check https://github.com/compomics/searchgui#troubleshooting and look for the individual search engines causing you trouble.

For Sage you may also want to try a different build than the default ones included in SearchGUI. You'll find all of the Sage builds for the latest release here: https://github.com/lazear/sage/releases/tag/v0.14.3.

Finally, note that the differences in the results between the different search engines are usually minor, hence we usually recommend only running one or a couple of them on your data. In other words, you do not need to run all of them.