[chimerax-users] Export pLDDT scores
Tom Goddard
goddard at sonic.net
Tue Nov 22 10:34:22 PST 2022
Hi Asmahan,
The pLDDT scores for an AlphaFold prediction are in the bfactor column of the PDB atomic model file. Here is an example of lines from such a file where the value 85.53 is the pLDDT for residue 1.
ATOM 1 N ASP B 1 -0.818 -36.026 -12.912 1.00 85.53 N
ATOM 2 CA ASP B 1 0.545 -35.783 -13.373 1.00 85.53 C
ATOM 3 C ASP B 1 0.706 -34.352 -13.882 1.00 85.53 C
The bfactor is assigned to every atom. But pLDDT is only assigned for each residue and measures CA position so you might want to look at only the CA lines.
$ grep CA ~/Downloads/ChimeraX/AlphaFold/prediction_14/af252_unrelaxed_rank_1_model_2.pdb
ATOM 2 CA ASP B 1 0.545 -35.783 -13.373 1.00 85.53 C
ATOM 10 CA CYS B 2 1.983 -32.130 -13.721 1.00 95.45 C
ATOM 16 CA PRO B 3 3.029 -30.770 -17.151 1.00 93.20 C
ATOM 23 CA SER B 4 6.629 -31.094 -18.066 1.00 93.36 C
You can also write the values into a ChimeraX attribute file using ChimeraX commands
select #1 at CA
save ~/Desktop/plddt.defattr attrName bfactor models #1 selectedOnly true
and that file looks like the following
attribute: bfactor
recipient: atoms
match mode: 1-to-1
/A:1 at CA 56.150001525878906
/A:2 at CA 53.33000183105469
/A:3 at CA 53.59000015258789
/A:4 at CA 59.380001068115234
documentation here
https://www.cgl.ucsf.edu/chimerax/docs/user/commands/save.html#attributes
Tom
> On Nov 22, 2022, at 7:08 AM, Asmahan Alghamdi <Asmahan.Alghamdi at nottingham.ac.uk> wrote:
>
> Dear Tom,
> I hope you are well. I was wondering if there is away to export pLDDT scores of the best predicted model of a protein using ChimeraX e.g. 1.5 (which I am using in my investigations)?
>
> Thank you very much.
> Asmahan
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.rbvi.ucsf.edu/pipermail/chimerax-users/attachments/20221122/80d415df/attachment.html>
More information about the ChimeraX-users
mailing list