[phenixbb] Alternative to AlphaFold2?

[Andy Watkins]

One significant advantage of OpenFold -- beyond its active open-source
development, its pytorch implementation (thus being more interoperable with
common external ML libraries) -- is its release of training code. So if
Odyssey wants to fine-tune predictions on a protein family of particular
interest to Odyssey, OpenFold's your answer. (You all can actually join the
OpenFold consortium -- which organizes not just OpenFold the existing
project but diverse additional open source ML development directions -- for
a small annual donation.)

In some respects single-sequence models like OmegaFold have advantages --
you don't need a heterogenous compute environment, with a good GPU for
inference but a lot of CPU cores for MSA precomputation. It's not a
total panacea though; embedding in a large language model is a significant
computation.

On Thu, Sep 22, 2022 at 11:09 AM Chip Lesburg <chip at odysseytx.com> wrote:

> Hi PHENIX users,
>
>
>
> At this past summer’s Diffraction Methods GRC meeting
> <https://www.grc.org/diffraction-methods-in-structural-biology-conference/2022/>,
> Tom Terwilliger presented the utility of incorporating AlphaFold2
> predictions during structure fitting and rebuilding. I wonder if anyone has
> experience using that or an alternative such as OmegaFold
> <https://www.biorxiv.org/content/10.1101/2022.07.21.500999v1> or OpenFold
> <https://github.com/aqlaboratory/openfold>? Are any of these easier or
> harder to get set up?
>
>
>
> Chip
>
>
> Charles Lesburg
>
> *Vice President, Structural Biology*
> *M* 908-265-2064
> 301 Binney Street, Cambridge, MA 02142
> *odysseytx.com <http://odysseytx.com>*
>
> This message is for the named person's use only. It may contain
> confidential, proprietary or legally privileged information. If you receive
> this message in error, please immediately delete it and all copies of it
> from your system, destroy any hard copies of it and notify the sender.
> ODYSSEY THERAPEUTICS, INC. ("ODYSSEY THERAPEUTICS") reserves the right to
> monitor all e-mail communications through its networks. ODYSSEY
> THERAPEUTICS cannot be committed by this e-mail or warrant the accuracy or
> validity of any information presented. Any views expressed in this message
> are those of the individual sender.
> _______________________________________________
> phenixbb mailing list
> phenixbb at phenix-online.org
> http://phenix-online.org/mailman/listinfo/phenixbb
> Unsubscribe: phenixbb-leave at phenix-online.org
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://phenix-online.org/pipermail/phenixbb/attachments/20220922/1944ce65/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 153156_edd5022b-de61-48a6-bf68-1a41de35ba89.png
Type: image/png
Size: 18009 bytes
Desc: not available
URL: <http://phenix-online.org/pipermail/phenixbb/attachments/20220922/1944ce65/attachment-0001.png>