ColabFold

Message boards : Rosetta@home Science : ColabFold

To post messages, you must log in.

AuthorMessage
Profile [VENETO] boboviz

Send message
Joined: 1 Dec 05
Posts: 1997
Credit: 9,747,451
RAC: 10,562
Message 102429 - Posted: 22 Aug 2021, 12:00:40 UTC

Interesting project: merge the best parts of AlphaFold2 and RosetTAFold in one solution!
And is open source!!
ID: 102429 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Sid Celery

Send message
Joined: 11 Feb 08
Posts: 2130
Credit: 41,424,155
RAC: 16,102
Message 102430 - Posted: 22 Aug 2021, 13:06:57 UTC - in response to Message 102429.  

ID: 102430 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Technologov

Send message
Joined: 17 Jul 07
Posts: 16
Credit: 415,434,481
RAC: 30,055
Message 102431 - Posted: 22 Aug 2021, 16:19:09 UTC

What if I don't have Google Colab, but I have a bunch of my own servers? (CPU+GPU)
Which code should I run?
ID: 102431 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Falconet

Send message
Joined: 9 Mar 09
Posts: 354
Credit: 1,252,725
RAC: 2,360
Message 102432 - Posted: 22 Aug 2021, 17:05:09 UTC - in response to Message 102431.  

What if I don't have Google Colab, but I have a bunch of my own servers? (CPU+GPU)
Which code should I run?



I suppose it would be this: https://github.com/RosettaCommons/RoseTTAFold
ID: 102432 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile [VENETO] boboviz

Send message
Joined: 1 Dec 05
Posts: 1997
Credit: 9,747,451
RAC: 10,562
Message 102433 - Posted: 22 Aug 2021, 18:03:13 UTC - in response to Message 102431.  

What if I don't have Google Colab, but I have a bunch of my own servers? (CPU+GPU)
Which code should I run?


My first link is to github source of the project...
ID: 102433 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile [VENETO] boboviz

Send message
Joined: 1 Dec 05
Posts: 1997
Credit: 9,747,451
RAC: 10,562
Message 102437 - Posted: 23 Aug 2021, 6:33:40 UTC - in response to Message 102430.  

And a nice presentation on it on youtube

ColabFold - Making protein folding accessible to all via Google Colab


Very interesting!!
ID: 102437 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile [VENETO] boboviz

Send message
Joined: 1 Dec 05
Posts: 1997
Credit: 9,747,451
RAC: 10,562
Message 102532 - Posted: 3 Sep 2021, 15:09:32 UTC

You can also run Colabfold locally on your pc: localcolabfold

....have an high-end gpu is better...
ID: 102532 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile [VENETO] boboviz

Send message
Joined: 1 Dec 05
Posts: 1997
Credit: 9,747,451
RAC: 10,562
Message 102874 - Posted: 2 Oct 2021, 17:07:51 UTC
Last modified: 2 Oct 2021, 17:11:24 UTC

Molecular simulation for everyone on Cloud with AlphaFold2, Amber and Charmm support
ID: 102874 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile [VENETO] boboviz

Send message
Joined: 1 Dec 05
Posts: 1997
Credit: 9,747,451
RAC: 10,562
Message 102875 - Posted: 2 Oct 2021, 17:08:49 UTC - in response to Message 102874.  

Local ColabFold now supports Cuda on Mac OSx
ID: 102875 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile [VENETO] boboviz

Send message
Joined: 1 Dec 05
Posts: 1997
Credit: 9,747,451
RAC: 10,562
Message 103059 - Posted: 30 Oct 2021, 14:29:27 UTC

Colabfold results

ColabFold offers accelerated protein structure and complex predictions by combining the fast homology search of MMseqs2 with AlphaFold2 or RoseTTAFold. ColabFold's 20-30x faster search and optimized model use allows predicting thousands of proteins per day on a server with one GPU

ID: 103059 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile [VENETO] boboviz

Send message
Joined: 1 Dec 05
Posts: 1997
Credit: 9,747,451
RAC: 10,562
Message 103166 - Posted: 11 Nov 2021, 14:56:09 UTC

Interesting news:

Alphafold2_mmseqs2 supports now to complex predictions using Alphafold-multimer.
+ Template predictions are temporary not possible in the Alphafold2_mmseqs2 notebook.
+ Advanced (Advaned_beta) notebooks still offer the old complex predictions logic.
+ ColabFold can be installed locally using pip!

ID: 103166 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile robbinespu

Send message
Joined: 2 Nov 21
Posts: 1
Credit: 11,533
RAC: 0
Message 103181 - Posted: 12 Nov 2021, 7:20:49 UTC - in response to Message 103166.  

that sound cool
ID: 103181 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile [VENETO] boboviz

Send message
Joined: 1 Dec 05
Posts: 1997
Credit: 9,747,451
RAC: 10,562
Message 105927 - Posted: 13 Apr 2022, 13:06:42 UTC
Last modified: 13 Apr 2022, 13:06:55 UTC

Some news from the project:

+ 11Mar2022 We use in default AlphaFold-multimer-v2 weights for complex modeling.
+ We also offer the old complex modes "AlphaFold-ptm" or "AlphaFold-multimer-v1"

+ 04Mar2022 ColabFold now uses a much more powerful server for MSAs and searches through the ColabFoldDB instead of BFD/MGnify.

ID: 105927 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile [VENETO] boboviz

Send message
Joined: 1 Dec 05
Posts: 1997
Credit: 9,747,451
RAC: 10,562
Message 106271 - Posted: 25 May 2022, 7:06:28 UTC - in response to Message 105927.  

ColabFold is now avaiable in COSMIC2 (web platform for submitting cryo-EM jobs through the cloud to Comet, a cluster at the San Diego Supercomputer Center)
ID: 106271 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile [VENETO] boboviz

Send message
Joined: 1 Dec 05
Posts: 1997
Credit: 9,747,451
RAC: 10,562
Message 106332 - Posted: 30 May 2022, 20:07:20 UTC

ID: 106332 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote
Profile [VENETO] boboviz

Send message
Joined: 1 Dec 05
Posts: 1997
Credit: 9,747,451
RAC: 10,562
Message 108033 - Posted: 8 Feb 2023, 21:08:02 UTC - in response to Message 106332.  

New version 1.5 (is integrating AlphaFold v2.3.1)
ID: 108033 · Rating: 0 · rate: Rate + / Rate - Report as offensive    Reply Quote

Message boards : Rosetta@home Science : ColabFold



©2024 University of Washington
https://www.bakerlab.org