Difference between revisions of "BinAC/Software/Alphafold"
F Bartusch (talk | contribs) (Created page with "{{Softwarepage|bio/alphafold}} {| width=600px class="wikitable" |- ! Description !! Content |- | module load | bio/alphafold |- | License | Apache License 2.0 - see [https://...") |
F Bartusch (talk | contribs) |
||
| Line 15: | Line 15: | ||
|- |
|- |
||
| Links |
| Links |
||
| − | | |
+ | | DeepMind AlphaFold Website: [https://deepmind.google/technologies/alphafold/] |
|} |
|} |
||
= Description = |
= Description = |
||
| + | AlphaFold developed by DeepMind predicts protein structures from the amino acid sequence at or near experimental resolution. |
||
| − | <br> |
||
| + | |||
| − | <br> |
||
| + | = Usage = |
||
| + | |||
| + | The BinAC provides Alphafold via an Apptainer Container. Both, the container and the AlphaFold database is stored on the WORK filesystem. |
||
| + | The module `bio/alphafold` provides a wrapper script called <code>alphafold</code>. |
||
| + | Upon loading the module, the wrapper <code>alphafold</code> is in <code>PATH</code> and can be directly used. |
||
| + | The wrapper behaves like the script used in DeepMind's AlphaFold GitHub repository. Thus, all options explained there are also applicable for our <code>alphafold</code> wrapper. |
||
= Parallel computing = |
= Parallel computing = |
||
| + | The first steps of AlphaFold compute three multiple sequence alignments (MSA). |
||
| − | = Usage = |
||
| + | These MSAs are computed on the CPU sequentially and the number of threads are hard-coded: |
||
| + | |||
| + | jackhmmer on UniRef90 using 8 threads</b> |
||
| + | jackhmmer on MGnify using 8 threads</b> |
||
| + | HHblits on BFD + Uniclust30 using 4 threads</b> |
||
| + | |||
| + | Thus, it does not make sense to use more than 8 cores in your job! |
||
| + | The additional cores will be idle. |
||
| + | |||
| + | The three MSAs are stored in the directory specified by <code>--output_dir</code> and can be reused with <code>--use_precomputed_msas=true</code>. |
||
| + | |||
| + | Only the model inference will run on the GPUs. |
||
Revision as of 15:27, 16 February 2024
|
The main documentation is available via |
| Description | Content |
|---|---|
| module load | bio/alphafold |
| License | Apache License 2.0 - see [1] |
| Citing | See [2] |
| Links | DeepMind AlphaFold Website: [3] |
1 Description
AlphaFold developed by DeepMind predicts protein structures from the amino acid sequence at or near experimental resolution.
2 Usage
The BinAC provides Alphafold via an Apptainer Container. Both, the container and the AlphaFold database is stored on the WORK filesystem.
The module `bio/alphafold` provides a wrapper script called alphafold.
Upon loading the module, the wrapper alphafold is in PATH and can be directly used.
The wrapper behaves like the script used in DeepMind's AlphaFold GitHub repository. Thus, all options explained there are also applicable for our alphafold wrapper.
3 Parallel computing
The first steps of AlphaFold compute three multiple sequence alignments (MSA). These MSAs are computed on the CPU sequentially and the number of threads are hard-coded:
jackhmmer on UniRef90 using 8 threads jackhmmer on MGnify using 8 threads HHblits on BFD + Uniclust30 using 4 threads
Thus, it does not make sense to use more than 8 cores in your job! The additional cores will be idle.
The three MSAs are stored in the directory specified by --output_dir and can be reused with --use_precomputed_msas=true.
Only the model inference will run on the GPUs.
