[buster-discuss] PDB Validate doesn't like mmCif file from Buster

[Marcin Wojdyr]

Hi Carlos,

Thanks for forwarding the discussion.
I also couldn't reproduce it. The lines that the PDB person advised to
remove have never caused any problems, so I'd guess it was a red
herring. Perhaps there was some intermittent problem that later went
away. If you'd use the validation server again, don't bother with
removing anything from mmcif. But of course if you'd see any issues
again we'd be happy to investigate it further.

As for running the validation locally, the pipelines that the PDB uses
for validation and deposition aren't fully open. I don't know if it'd
be possible to reproduce them. Some older programs used in these
pipelines are available on the RCSB website and occasionally new
versions are uploaded there. There are also some programs maintained
in Github repositories. However, that's probably not all the software
that is used and I don't have a complete picture of what is running on
the PDB servers.

The issue you reported was from mid October. Coincidentally, a week or
two before that I reported an unrelated issue to the PDB. In response,
one of the PDB developers released a new version of MAXIT (the program
that the PDB uses to read and convert coordinate files). That was on
Oct 11. Probably the validation service was at some point updated to
use that new version, but according to another PDB person OneDep
hasn't been updated yet. Tracking such problems often feels like
navigating in the dark.

Best wishes,
Marcin

On Wed, Dec 4, 2024 at 1:20 PM Kikuti Carlos via buster-discuss
<buster-discuss at globalphasing.com> wrote:
>
> Hello !
>
>
>
> Marcin, I’ve just sent it to your e-mail, thank you.
>
>
>
> If you or anybody else has ideas concerning the second part: “To ask if you have other suggestions of routines, other than uploading to Validate for this kind of pre-submission test? We use the validation tools in Coot and Molprobity, but for some reason the criteria and the calculated maps are not exactly the same as for the PDB… It would be much easier to have this validation running locally, just to avoid the uploading interface.”
>
> … I’m all ears.
>
>
>
> Best,
>
>
>
> ---------------------------------------------------
> Carlos KIKUTI, PhD
> UMR144 - CNRS - Institut Curie
>
> Pavillon Trouillet Rossignol
> 26 Rue d’Ulm - 75005 Paris, France
> carlos.kikuti at curie.fr
>
>
> Message: 2
> Date: Tue, 3 Dec 2024 17:57:58 +0100
> From: Marcin Wojdyr <wojdyr at gmail.com>
> To: Kikuti Carlos <Carlos.Kikuti at curie.fr>
> Cc: "buster-discuss at globalphasing.com"
>         <buster-discuss at globalphasing.com>, Glaser Camille
>         <camille.glaser at curie.fr>
> Subject: Re: [buster-discuss] PDB Validate doesn't like mmCif file
>         from    Buster
> Message-ID:
>         <CACaHzQXvKWhbkFHYUvz4ZFqBnwhABALrovAKWTawK8+dWs4K-A at mail.gmail.com>
> Content-Type: text/plain; charset="UTF-8"
>
> Hello Carlos,
>
> Thanks for reporting this problem. I don't know what went wrong here.
> It'd be good to track it down, but I'm not able to reproduce it.
>
> How did you discuss it with the PDB people? If by email, could you
> forward it to me?
>
> Thanks,
> Marcin
> *********************************************
>
> _______________________________________________
> buster-discuss mailing list
> buster-discuss at globalphasing.com
> https://www.globalphasing.com/mailman/listinfo/buster-discuss