Rosetta Ignores .flag file when using .tar as Input

[narek-abelyan]

Hi!
I noticed some strange behavior when running experiments using Rosetta and OpenVS.

You are using the @flag_path option, which points to a .flag file. This file seems to refer to .tar archives, but not to .params files, (which are needed to work properly with ligands). However, it looks like Rosetta does not handle .tar files correctly or maybe doesn't support them at all as input. Because of that, it ignores the flags and only docks the reference ligand, without using other important input data.

This makes it impossible to reproduce the experiment - the results are incomplete or incorrect.

Maybe it’s just a small mistake in the .flag file structure, and it could be fixed by changing how the files point to each other. Or maybe I misunderstood how the setup is supposed to work.

I’d really appreciate it if you could clarify how to set things up so that Rosetta uses all the necessary parameters correctly.

[xizhihui]

@narek-abelyan Hi,
Thank you very much for the comment. I changed the flag that you mentioned into multiple "key:value" pairs, and finally got the right number of scores and ligand names . That's great !!
Did you run through the example E3L_6DO3 or the other one? Any suggestion to run it on our own data?