Skip to content

Commit

Permalink
Update the disclaimer to match AlphaFold 3 + run mdformat.
Browse files Browse the repository at this point in the history 
PiperOrigin-RevId: 696460542
Change-Id: I42defc0a687207ee57801b0aeba78fe1e63b03e6
  • Loading branch information
@Augustin-Zidek @copybara-github
Augustin-Zidek authored and copybara-github committed Nov 14, 2024
1 parent 85e0be3 commit d95a92a
Showing 1 changed file with 45 additions and 34 deletions.
79 changes: 45 additions & 34 deletions README.md
View file
Original file line number Diff line number Diff line change
Expand Up @@ -46,6 +46,7 @@ with more memory can predict larger protein structures).
Please follow these steps:

1. Install [Docker](https://www.docker.com/).

* Install
[NVIDIA Container Toolkit](https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/install-guide.html)
for GPU support.
Expand All @@ -62,25 +63,24 @@ Please follow these steps:
1. Download genetic databases and model parameters:

* Install `aria2c`. On most Linux distributions it is available via the
package manager as the `aria2` package (on Debian-based distributions this
can be installed by running `sudo apt install aria2`).
package manager as the `aria2` package (on Debian-based distributions
this can be installed by running `sudo apt install aria2`).

* Please use the script `scripts/download_all_data.sh` to download
and set up full databases. This may take substantial time (download size is
556 GB), so we recommend running this script in the background:
* Please use the script `scripts/download_all_data.sh` to download and set
up full databases. This may take substantial time (download size is 556
GB), so we recommend running this script in the background:

```bash
scripts/download_all_data.sh <DOWNLOAD_DIR> > download.log 2> download_all.log &
```

* **Note: The download directory `<DOWNLOAD_DIR>` should *not* be a
subdirectory in the AlphaFold repository directory.** If it is, the Docker
build will be slow as the large databases will be copied into the docker
build context.
subdirectory in the AlphaFold repository directory.** If it is, the
Docker build will be slow as the large databases will be copied into the
docker build context.

* It is possible to run AlphaFold with reduced databases; please refer to
the [complete documentation](#genetic-databases).

the [complete documentation](#genetic-databases).

1. Check that AlphaFold will be able to use a GPU by running:

Expand All @@ -95,8 +95,9 @@ Please follow these steps:
[NVIDIA Docker issue](https://github.com/NVIDIA/nvidia-docker/issues/1447#issuecomment-801479573).
If you wish to run AlphaFold using Singularity (a common containerization
platform on HPC systems) we recommend using some of the third party Singularity
setups as linked in https://github.com/deepmind/alphafold/issues/10 or
platform on HPC systems) we recommend using some of the third party
Singularity setups as linked in
https://github.com/deepmind/alphafold/issues/10 or
https://github.com/deepmind/alphafold/issues/24.
1. Build the Docker image:
Expand Down Expand Up @@ -312,7 +313,8 @@ To use the deprecated v2.1.0 AlphaFold-Multimer model weights:
was tested on Google Cloud with a machine using the `nvidia-gpu-cloud-image`
with 12 vCPUs, 85 GB of RAM, a 100 GB boot disk, the databases on an additional
3 TB disk, and an A100 GPU. For your first run, please follow the instructions
from [Installation and running your first prediction](#installation-and-running-your-first-prediction)
from
[Installation and running your first prediction](#installation-and-running-your-first-prediction)
section.
1. By default, Alphafold will attempt to use all visible GPU devices. To use a
Expand Down Expand Up @@ -363,15 +365,15 @@ section.
--output_dir=/home/user/absolute_path_to_the_output_dir
```
1. After generating the predicted model, AlphaFold runs a relaxation
step to improve local geometry. By default, only the best model (by
pLDDT) is relaxed (`--models_to_relax=best`), but also all of the models
1. After generating the predicted model, AlphaFold runs a relaxation step to
improve local geometry. By default, only the best model (by pLDDT) is
relaxed (`--models_to_relax=best`), but also all of the models
(`--models_to_relax=all`) or none of the models (`--models_to_relax=none`)
can be relaxed.
1. The relaxation step can be run on GPU (faster, but could be less stable) or
CPU (slow, but stable). This can be controlled with `--enable_gpu_relax=true`
(default) or `--enable_gpu_relax=false`.
CPU (slow, but stable). This can be controlled with
`--enable_gpu_relax=true` (default) or `--enable_gpu_relax=false`.
1. AlphaFold can re-use MSAs (multiple sequence alignments) for the same
sequence via `--use_precomputed_msas=true` option; this can be useful for
Expand Down Expand Up @@ -405,12 +407,12 @@ can be done via the `--num_multimer_predictions_per_model` flag, e.g. set it to
### AlphaFold prediction speed
The table below reports prediction runtimes for proteins of various lengths. We
only measure unrelaxed structure prediction with three recycles while
excluding runtimes from MSA and template search. When running
`docker/run_docker.py` with `--benchmark=true`, this runtime is stored in
`timings.json`. All runtimes are from a single A100 NVIDIA GPU. Prediction
speed on A100 for smaller structures can be improved by increasing
`global_config.subbatch_size` in `alphafold/model/config.py`.
only measure unrelaxed structure prediction with three recycles while excluding
runtimes from MSA and template search. When running `docker/run_docker.py` with
`--benchmark=true`, this runtime is stored in `timings.json`. All runtimes are
from a single A100 NVIDIA GPU. Prediction speed on A100 for smaller structures
can be improved by increasing `global_config.subbatch_size` in
`alphafold/model/config.py`.
No. residues | Prediction time (s)
-----------: | ------------------:
Expand Down Expand Up @@ -577,14 +579,14 @@ The contents of each output file are as follows:
structure prediction (see Jumper et al. 2021, Suppl. Methods 1.8.6 for
details).
* `ranked_*.pdb` – A PDB format text file containing the predicted structures,
after reordering by model confidence. Here `ranked_i.pdb` should contain
the prediction with the (`i + 1`)-th highest confidence (so that
`ranked_0.pdb` has the highest confidence). To rank model confidence, we use
predicted LDDT (pLDDT) scores (see Jumper et al. 2021, Suppl. Methods 1.9.6
for details). If `--models_to_relax=all` then all ranked structures are
relaxed. If `--models_to_relax=best` then only `ranked_0.pdb` is relaxed
(the rest are unrelaxed). If `--models_to_relax=none`, then the ranked
structures are all unrelaxed.
after reordering by model confidence. Here `ranked_i.pdb` should contain the
prediction with the (`i + 1`)-th highest confidence (so that `ranked_0.pdb`
has the highest confidence). To rank model confidence, we use predicted LDDT
(pLDDT) scores (see Jumper et al. 2021, Suppl. Methods 1.9.6 for details).
If `--models_to_relax=all` then all ranked structures are relaxed. If
`--models_to_relax=best` then only `ranked_0.pdb` is relaxed (the rest are
unrelaxed). If `--models_to_relax=none`, then the ranked structures are all
unrelaxed.
* `ranking_debug.json` – A JSON format text file containing the pLDDT values
used to perform the model ranking, and a mapping back to the original model
names.
Expand Down Expand Up @@ -693,7 +695,6 @@ If you use the code or data in this package, please cite:
In addition, if you use the AlphaFold-Multimer mode, please cite:
```bibtex
@article {AlphaFold-Multimer2021,
author = {Evans, Richard and O{\textquoteright}Neill, Michael and Pritzel, Alexander and Antropova, Natasha and Senior, Andrew and Green, Tim and {\v{Z}}{\'\i}dek, Augustin and Bates, Russ and Blackwell, Sam and Yim, Jason and Ronneberger, Olaf and Bodenstein, Sebastian and Zielinski, Michal and Bridgland, Alex and Potapenko, Anna and Cowie, Andrew and Tunyasuvunakool, Kathryn and Jain, Rishub and Clancy, Ellen and Kohli, Pushmeet and Jumper, John and Hassabis, Demis},
Expand Down Expand Up @@ -765,6 +766,17 @@ This is not an officially supported Google product.
Copyright 2022 DeepMind Technologies Limited.
AlphaFold 2 and its output are for theoretical modeling only. They are not
intended, validated, or approved for clinical use. You should not use the
AlphaFold 2 or its output for clinical purposes or rely on them for medical or
other professional advice. Any content regarding those topics is provided for
informational purposes only and is not a substitute for advice from a qualified
professional.
Output of AlphaFold 2 are predictions with varying levels of confidence and
should be interpreted carefully. Use discretion before relying on, publishing,
downloading or otherwise using AlphaFold 2 and its output.
### AlphaFold Code License
Licensed under the Apache License, Version 2.0 (the "License"); you may not use
Expand Down Expand Up @@ -816,4 +828,3 @@ reference to the following:
restrictions and made fully and freely available for both non-commercial and
commercial use under
[CC0 1.0 Universal (CC0 1.0) Public Domain Dedication](https://creativecommons.org/publicdomain/zero/1.0/).

0 comments on commit d95a92a

Please sign in to comment.