[chimerax-users] AlphaFold best_model.pdb could not visually overlap the input sequence using matchmaker

[Tristan Croll]

Yes - I just had the models opened in the opposite order to you.

Ultimately, it seems to me that the problem here is that you’re asking for
something impossible: a good overall structural alignment between two
models that are fundamentally different. A rigid body alignment can only
rotate and translate one model to overlay the other. If the two models are
in very different conformations you have two choices: (a) get a bad overall
all-atom alignment, or (b) try to find the biggest piece that *does* align
well and just align that. By default MatchMaker takes the latter approach,
which is why when you use it on your two models it ends up aligning just
the antigen-binding region of the antibody fragment at the expense of
everything else - that’s the biggest part of the two models that *can* be
aligned well.

If you really want approach (a) you could use MatchMaker with the argument
“cutOff None”, in which case it will try to find the best-aligning whole
chain and align on that. You can also specify exactly which pair of chains
you want to align on, as per the example in my previous email. Ultimately,
though, nothing you do here is going to turn a bad AlphaFold prediction
into a good one.

— Tristan

On Sun, 26 Mar 2023 at 22:43, Cheng Fei Phung <feiphung at hotmail.com> wrote:

> Hi Tristan Croll,
>
> I presume that you meant "match #2/H to #1/D" and "match #2/L to #1/C"
> instead?
>
> See https://i.imgur.com/dXCzuIP.png and https://i.imgur.com/VbKhKCd.png
>
> Regards,
> Phung Cheng Fei
> ------------------------------
> *From:* Tristan Croll <tcroll at altoslabs.com>
> *Sent:* Monday, March 27, 2023 1:10 AM
>
> *To:* Cheng Fei Phung <feiphung at hotmail.com>
> *Cc:* chimerax-users at cgl.ucsf.edu <chimerax-users at cgl.ucsf.edu>
> *Subject:* Re: [chimerax-users] AlphaFold best_model.pdb could not
> visually overlap the input sequence using matchmaker
>
> ... huh. Looking at it again, it seems to me that AlphaFold has predicted
> the antibody heavy chain in the light chain conformation and vice versa. If
> you force it to align (predicted heavy) to (experimental light) with "match
> #2/D to #1/H" then all four antibody domains align reasonably well. Seems
> like a failure on AlphaFold's part (and one I haven't seen before).
>
> On Sun, Mar 26, 2023 at 1:37 PM Cheng Fei Phung via ChimeraX-users <
> chimerax-users at cgl.ucsf.edu> wrote:
>
> Hi,
>
> I am trying do "matchmaker #2 to #1" between colab output ( best_model.pdb
> which is pasted here at http://ix.io/4rRO ) and
> https://www.rcsb.org/structure/7CR5
>
> However, as seen in https://i.imgur.com/gHL8to3.png , there is not really
> much visual overlapping since the structures are not rotated properly to
> match each other.
>
> The log for doing matchmaker command is available here at
> https://i.imgur.com/78h5CLq.png
>
> The colab notebook is also saved as a screenshot here at
> https://i.imgur.com/LSTdyZe.png
>
> Manually doing "turn z 1 270 model #1 center #1" or similar rotations
> commands also do not help.
> Please advise.
>
> Regards,
> Phung Cheng Fei
> _______________________________________________
> ChimeraX-users mailing list
> ChimeraX-users at cgl.ucsf.edu
> Manage subscription:
> https://www.rbvi.ucsf.edu/mailman/listinfo/chimerax-users
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.rbvi.ucsf.edu/pipermail/chimerax-users/attachments/20230327/00bc4646/attachment.html>