examples/alphafold2.md
... ...
@@ -1,8 +1,18 @@
1 1
2 2
## ALPHAFOLD2
3
-
3
+<!-- TOC -->
4
+
5
+- [ALPHAFOLD2](#alphafold2)
6
+ - [Preparing to run Alphafold](#preparing-to-run-alphafold)
7
+ - [Using the default run_alphafold.sh wrapper](#using-the-default-run_alphafoldsh-wrapper)
8
+ - [Creating your own run_alphafold script](#creating-your-own-run_alphafold-script)
9
+ - [Running the python script run_alphafold.py directly](#running-the-python-script-run_alphafoldpy-directly)
10
+ - [Examples:](#examples)
11
+ - [Web portal](#web-portal)
12
+ - [Known issues](#known-issues)
13
+
14
+<!-- /TOC -->
4 15
### Preparing to run Alphafold
5
-
6 16
The ALPHAFOLD2 source an implementation of the inference pipeline of AlphaFold v2.0. using a completely new model that was entered in CASP14. This is not a production application per se, but a reference that is capable of producing structures from a single amino acid sequence.
7 17
8 18
The SBGrid installation of Alphafold2 does not require Docker to run, but does require a relatively recent NVidia GPU and updated driver.
... ...
@@ -59,6 +69,7 @@ The database directory shouuld look like this :
59 69
└── uniref90.fasta
60 70
```
61 71
72
+### Using the default run_alphafold.sh wrapper
62 73
Once the databases are in place, AlphaFold can be run with the wrapper script run_alphafold.sh. The default location for the databases should be `/programs/local/alphafold`, but can be changed using the ALPHAFOLD_DB variable. For example:
63 74
64 75
```
... ...
@@ -78,10 +89,27 @@ To use the run script, specify the path to the fasta file and an output directy
78 89
run_alphafold.sh <path to fasta file> <path to an output directory>
79 90
```
80 91
92
+### Creating your own run_alphafold script
93
+
94
+You can use our run_alphafold script template here to create your own run script.
95
+
96
+[run_alphafold_template.sh](run_alphfold_template.sh)
97
+
98
+### Running the python script run_alphafold.py directly
81 99
run_alphafold.sh is a convenience wrapper script that shortens the required command arguments to run_alphafold.py. The run_alphafold.py script is also available which requires all parameters to be set explicitly, but provides greater flexibility. Pass --helpshort or --helpfull to see help on flags.
82 100
101
+### Examples:
102
+
103
+We include reference sequences from CASP14 in the installation.
104
+This command should run successfully:
105
+
106
+```
107
+run_alphafold.sh /programs/x86_64-linux/alphafold/2.0.0/alphafold/data/T1050.fasta
108
+```
83 109
110
+### Web portal
84 111
It is possible to run alphafold through a web portal. See
85 112
https://colab.research.google.com/github/deepmind/alphafold/blob/main/notebooks/AlphaFold.ipynb .
86 113
114
+### Known issues
87 115
Known issues: This version may not run on some newer GPUs that require CUDA 11.1 or later. An update is coming that should correct this.
... ...
\ No newline at end of file
examples/run_alphafold_template.sh
... ...
@@ -0,0 +1,57 @@
1
+#!/bin/bash
2
+# Jason Key 1fac4ac 2021-07-23 08:33:47 -0400
3
+# Copyright © 2021 SBGrid Consortium. All rights reserved.
4
+#
5
+# wrapper script for alphafold
6
+
7
+USAGE="$(basename $0): A wrapper script for running Alphafold jobs in the SBGrid Software installation \n\n
8
+Usage: $(basename $0) [fasta file] [path to output directory] \n\n"
9
+
10
+if [ $# -eq 0 ]; then
11
+ echo -e $USAGE
12
+ exit 1
13
+fi
14
+
15
+## To set single GPU
16
+# export CUDA_VISIBLE_DEVICES=0
17
+
18
+data_dir="/programs/local/alphafold/"
19
+if [ ! -z ${ALPHAFOLD_DB} ] && [ -d ${ALPHAFOLD_DB} ] ; then
20
+ data_dir=${ALPHAFOLD_DB}
21
+fi
22
+
23
+if [ ! -d ${data_dir} ] ; then
24
+ echo "${data_dir} is not a directory. Exiting... "
25
+ exit 1
26
+fi
27
+
28
+echo "Using databases in ${data_dir}"
29
+
30
+input_fasta=$1
31
+
32
+if [ ! -f ${input_fasta} ] ; then
33
+ echo "${input_fasta} is not a file. Exiting... "
34
+ exit 1
35
+fi
36
+
37
+output_dir="/tmp/alphafold"
38
+if [ $2 ] ; then
39
+ output_dir=$2
40
+fi
41
+
42
+mkdir -p "${output_dir}"
43
+
44
+/programs/x86_64-linux/alphafold/2.0.0/bin.capsules/run_alphafold.py \
45
+--model_names="model_1,model_2,model_3,model_4,model_5" \
46
+--data_dir="${data_dir}" \
47
+--uniref90_database_path="${data_dir}/uniref90/uniref90.fasta" \
48
+--mgnify_database_path="${data_dir}/mgnify/mgy_clusters.fa" \
49
+--uniclust30_database_path="${data_dir}/uniclust30/uniclust30_2018_08/uniclust30_2018_08" \
50
+--bfd_database_path="${data_dir}/bfd/bfd_metaclust_clu_complete_id30_c90_final_seq.sorted_opt" \
51
+--pdb70_database_path="${data_dir}/pdb70/pdb70" \
52
+--template_mmcif_dir="${data_dir}/pdb_mmcif/mmcif_files" \
53
+--obsolete_pdbs_path="$data_dir/pdb_mmcif/obsolete.dat" \
54
+--max_template_date="2020-05-14" \
55
+--preset="full_dbs" \
56
+--output_dir="${output_dir}" \
57
+--fasta_paths="${input_fasta}"