{bio}[foss/2020b] AlphaFold v2.0.0 w/ Python 3.8.6

easybuild/easyconfigs/a/AlphaFold/AlphaFold-2.0.0-foss-2020b.eb

@@ -0,0 +1,174 @@
+easyblock = 'PythonBundle'
+
+name = 'AlphaFold'
+version = '2.0.0'
+
+homepage = 'https://deepmind.com/research/case-studies/alphafold'
+description = "AlphaFold can predict protein structures with atomic accuracy even where no similar structure is known"
+
+toolchain = {'name': 'foss', 'version': '2020b'}
+
+builddependencies = [
+    # CMake/oxygen/SWIG are required for building OpenMM
+    ('CMake', '3.18.4'),
+    ('Doxygen', '1.8.20'),
+    # required for installing dm-tree
+    ('Bazel', '3.7.2'),
+    ('SWIG', '4.0.2'),
+]
+
+dependencies = [
+    ('Python', '3.8.6'),
+    ('SciPy-bundle', '2020.11'),
+    ('PyYAML', '5.3.1'),
+    ('TensorFlow', '2.4.1'),
+    ('Biopython', '1.78'),
+    ('HH-suite', '3.3.0'),
+    ('HMMER', '3.3.2'),
+    ('Kalign', '3.3.1'),
+    ('PDBFixer', '1.7'),
+    ('jax', '0.2.19'),  # also provides absl-py
+]
+
+local_openmm_preinstallopts = "export OPENMM_INCLUDE_PATH=%(installdir)s/include && "
+local_openmm_preinstallopts += " export OPENMM_LIB_PATH=%(installdir)s/lib && "
+
+# required to install OpenMM Python API;
+# avoid that setup.py partially uninstalls existing OpenMM Python bindings...
+local_openmm_installopts = " && cd python && sed -i 's/uninstall()/pass/g' setup.py && "
+local_openmm_installopts += "python setup.py build && python setup.py install --prefix=%(installdir)s"
+
+# commit to use for downloading stereo_chemical_props.txt and copy to alphafold/common,
+# see docker/Dockerfile in AlphaFold repository
+local_scp_commit = '7102c6'
+
+components = [
+    # for simtk
+    ('OpenMM', '7.5.1', {
+        'easyblock': 'CMakeMake',
+        'source_urls': ['https://github.com/openmm/openmm/archive/'],
+        'sources': [{'download_filename': '%(version)s.tar.gz', 'filename': SOURCE_TAR_GZ}],
+        'patches': [('OpenMM-%(version)s_AlphaFold.patch', 'wrappers/python')],
+        'checksums': [
+            'c88d6946468a2bde2619acb834f57b859b5e114a93093cf562165612e10f4ff7',  # OpenMM-7.5.1.tar.gz
+            '1b109dfff3af5c6aa70690bca14618612953c68840a7e64f679db7ca33c1aff6',  # OpenMM-7.5.1_AlphaFold.patch
+        ],
+        'start_dir': 'openmm-%(version)s',
+        'preinstallopts': local_openmm_preinstallopts,
+        'installopts': local_openmm_installopts,
+    }),
+    (name, version, {
+        'easyblock': 'PythonPackage',
+        'source_urls': [
+            'https://github.com/deepmind/alphafold/archive/refs/tags/',
+            'https://git.scicore.unibas.ch/schwede/openstructure/-/raw/%s/modules/mol/alg/src/' % local_scp_commit,
+        ],
+        'sources': [
+            {
+                'download_filename': 'v%(version)s.tar.gz',
+                'filename': SOURCE_TAR_GZ,
+            },
+            {
+                'download_filename': 'stereo_chemical_props.txt',
+                'filename': 'stereo_chemical_props-%s.txt' % local_scp_commit,
+                'extract_cmd': "cp -a %s .",
+            },
+        ],
+        'patches': [
+            'AlphaFold-%(version)s_fix-packages.patch',
+            'AlphaFold-%(version)s_data-dep-paths.patch',
+            'AlphaFold-%(version)s_n-cpu.patch',
+            'AlphaFold-2.0.0_fix-scp-path.patch',
+        ],
+        'checksums': [
+            '78cf443a2d9250917b05f5d40ede140ed8e8341b42fdfa54340336aca52f53f2',  # v2.0.0.tar.gz
+            '24510899eeb49167cffedec8fa45363a4d08279c0c637a403b452f7d0ac09451',  # stereo_chemical_props-7102c6.txt
+            '826d2d1a5d6ac52c51a60ba210e1947d5631a1e2d76f8815305b5d23f74458db',  # AlphaFold-2.0.0_fix-packages.patch
+            '7223e297b23f90816219095696bc6453910c617add60b907a0d3e869da8733bb',  # AlphaFold-2.0.0_data-dep-paths.patch
+            'dfda4dd5f9aba19fe2b6eb9a0ec583d12dcefdfee8ab8803fc57ad48d582db04',  # AlphaFold-2.0.0_n-cpu.patch
+            '83dc82a8b1c647eb7e217aef683153e98a4fc7f871a85280976c92a1bfe28f27',  # AlphaFold-2.0.0_fix-scp-path.patch
+        ],
+        'start_dir': 'alphafold-%(version)s',
+        'use_pip': True,
+    }),
+]
+
+use_pip = True
+
+exts_list = [
+    ('toolz', '0.11.1', {
+        'checksums': ['c7a47921f07822fe534fb1c01c9931ab335a4390c782bd28c6bcc7c2f71f3fbf'],
+    }),
+    ('chex', '0.0.8', {
+        'checksums': ['d6ce1329470116b6f172a72b1131bfd4d11fb7eb465e6077c3b36224b5b09fe4'],
+    }),
+    ('tabulate', '0.8.9', {
+        'checksums': ['eb1d13f25760052e8931f2ef80aaf6045a6cceb47514db8beab24cded16f13a7'],
+    }),
+    ('dm-haiku', '0.0.4', {
+        # source tarball on PyPI is missing requirements.txt,
+        # see https://github.com/deepmind/dm-haiku/issues/44
+        'source_urls': ['https://github.com/deepmind/dm-haiku/archive/refs/tags/'],
+        'sources': [{'download_filename': 'v%(version)s.tar.gz', 'filename': SOURCE_TAR_GZ}],
+        'checksums': ['e9896d161938b53f869fde207f3f6fca496d09b2a47d21dd9b6b65e897ad6aab'],
+        'modulename': 'haiku',
+    }),
+    ('dm-tree', '0.1.6', {
+        'checksums': ['6776404b23b4522c01012ffb314632aba092c9541577004ab153321e87da439a'],
+        'modulename': 'tree',
+    }),
+    ('websocket-client', '1.2.1', {
+        'checksums': ['8dfb715d8a992f5712fff8c843adae94e22b22a99b2c5e6b0ec4a1a981cc4e0d'],
+        'modulename': 'websocket',
+    }),
+    ('docker', '5.0.0', {
+        'checksums': ['3e8bc47534e0ca9331d72c32f2881bb13b93ded0bcdeab3c833fb7cf61c0a9a5'],
+    }),
+    ('immutabledict', '2.1.0', {
+        'checksums': ['673fb8f30f46d23dd394050b979f5b7f4c5398982b99ebc854fb873e646b967a'],
+    }),
+    ('contextlib2', '21.6.0', {
+        'checksums': ['ab1e2bfe1d01d968e1b7e8d9023bc51ef3509bba217bb730cee3827e1ee82869'],
+    }),
+    ('ml_collections', '0.1.0', {
+        'checksums': ['59a17fcd1c140153009788517f304caaddd7a94f06690f9f0ed09987beebcf3c'],
+        # see https://github.com/google/ml_collections/issues/7
+        'preinstallopts': "touch requirements.txt && touch requirements-test.txt && ",
+    }),
+]
+
+postinstallcmds = [
+    "mkdir -p %(installdir)s/bin",
+    # run_alphafold.py script is missing a shebang...
+    "echo '#!/usr/bin/env python' > %(installdir)s/bin/run_alphafold.py",
+    "cat %(builddir)s/alphafold-%(version)s/run_alphafold.py >> %(installdir)s/bin/run_alphafold.py",
+    "chmod a+x %(installdir)s/bin/run_alphafold.py",
+    "cd %(installdir)s/bin && ln -s run_alphafold.py alphafold",
+    "cp -a %(builddir)s/alphafold-%(version)s/scripts %(installdir)s/",
+    "cp -a %%(builddir)s/stereo_chemical_props-%s.txt %%(installdir)s/stereo_chemical_props.txt" % local_scp_commit,
+    # run tests for run_alphafold.py script;
+    # shouldn't do this in sanity check to avoid breaking use of --module-only
+    "PYTHONPATH=%(installdir)s/lib/python%(pyshortver)s/site-packages:$PYTHONPATH "
+    "python %(builddir)s/alphafold-%(version)s/run_alphafold_test.py",
+]
+
+sanity_check_paths = {
+    'files': ['bin/alphafold', 'bin/run_alphafold.py', 'lib/libOpenMM.%s' % SHLIB_EXT,
+              'lib/python%(pyshortver)s/site-packages/simtk/openmm/openmm.py', 'stereo_chemical_props.txt'],
+    'dirs': ['lib/python%(pyshortver)s/site-packages', 'scripts'],
+}
+
+sanity_check_commands = [
+    "python -m simtk.testInstallation",
+    "python -c 'import alphafold'",
+    "alphafold --help 2>&1 | grep 'Full AlphaFold protein structure prediction script'",
+]
+
+sanity_pip_check = True
+
+modextrapaths = {
+    'OPENMM_INCLUDE_PATH': 'include',
+    'OPENMM_LIB_PATH': 'lib',
+}
+
+moduleclass = 'bio'

easybuild/easyconfigs/a/AlphaFold/AlphaFold-2.0.0_data-dep-paths.patch

@@ -0,0 +1,115 @@
+pick up on $ALPHAFOLD_DATA_DIR to specify location to downloaded data
+(see https://github.com/deepmind/alphafold/blob/main/docker/run_docker.py);
+pick up on HH-suite, HHMER, Kalign dependencies provided via EasyBuild
+author: Kenneth Hoste (HPC-UGent)
+diff --git a/run_alphafold.py b/run_alphafold.py
+index 61f7ef2..81c97a0 100644
+--- a/run_alphafold.py
++++ b/run_alphafold.py
+@@ -36,6 +36,46 @@ from alphafold.model import model
+ from alphafold.relax import relax
+ # Internal import (7716).
+
++
++data_dir = os.getenv('ALPHAFOLD_DATA_DIR')
++if data_dir:
++    uniref90_database_path = os.path.join(data_dir, 'uniref90', 'uniref90.fasta')
++    mgnify_database_path = os.path.join(data_dir, 'mgnify', 'mgy_clusters.fa')
++    bfd_database_path = os.path.join(data_dir, 'bfd', 'bfd_metaclust_clu_complete_id30_c90_final_seq.sorted_opt')
++    uniclust30_database_path = os.path.join(data_dir, 'uniclust30', 'uniclust30_2018_08', 'uniclust30_2018_08')
++    pdb70_database_path = os.path.join(data_dir, 'pdb70', 'pdb70')
++    template_mmcif_dir = os.path.join(data_dir, 'pdb_mmcif', 'mmcif_files')
++    obsolete_pdbs_path = os.path.join(data_dir, 'pdb_mmcif', 'obsolete.dat')
++else:
++    sys.stderr.write("$ALPHAFOLD_DATA_DIR is not defined!")
++    uniref90_database_path = None
++    mgnify_database_path = None
++    bfd_database_path = None
++    uniclust30_database_path = None
++    pdb70_database_path = None
++    template_mmcif_dir = None
++    obsolete_pdbs_path = None
++
++hmmer_root = os.getenv('EBROOTHMMER')
++if hmmer_root:
++    jackhmmer_binary_path = os.path.join(hmmer_root, 'bin', 'jackhmmer')
++else:
++    jackhmmer_binary_path = '/usr/bin/jackhmmer'
++
++hhsuite_root = os.getenv('EBROOTHHMINSUITE')
++if hhsuite_root:
++    hhblits_binary_path = os.path.join(hhsuite_root, 'bin', 'hhblits')
++    hhsearch_binary_path = os.path.join(hhsuite_root, 'bin', 'hhsearch')
++else:
++    hhblits_binary_path = '/usr/bin/hhblits'
++    hhsearch_binary_path = '/usr/bin/hhsearch'
++
++kalign_root = os.getenv('EBROOTKALIGN')
++if kalign_root:
++    kalign_binary_path = os.path.join(kalign_root, 'bin', 'kalign')
++else:
++    kalign_binary_path = '/usr/bin/kalign'
++
+ flags.DEFINE_list('fasta_paths', None, 'Paths to FASTA files, each containing '
+                   'one sequence. Paths should be separated by commas. '
+                   'All FASTA paths must have a unique basename as the '
+@@ -44,30 +83,30 @@ flags.DEFINE_list('fasta_paths', None, 'Paths to FASTA files, each containing '
+ flags.DEFINE_string('output_dir', None, 'Path to a directory that will '
+                     'store the results.')
+ flags.DEFINE_list('model_names', None, 'Names of models to use.')
+-flags.DEFINE_string('data_dir', None, 'Path to directory of supporting data.')
+-flags.DEFINE_string('jackhmmer_binary_path', '/usr/bin/jackhmmer',
++flags.DEFINE_string('data_dir', data_dir, 'Path to directory of supporting data.')
++flags.DEFINE_string('jackhmmer_binary_path', jackhmmer_binary_path,
+                     'Path to the JackHMMER executable.')
+-flags.DEFINE_string('hhblits_binary_path', '/usr/bin/hhblits',
++flags.DEFINE_string('hhblits_binary_path', hhblits_binary_path,
+                     'Path to the HHblits executable.')
+-flags.DEFINE_string('hhsearch_binary_path', '/usr/bin/hhsearch',
++flags.DEFINE_string('hhsearch_binary_path', hhsearch_binary_path,
+                     'Path to the HHsearch executable.')
+-flags.DEFINE_string('kalign_binary_path', '/usr/bin/kalign',
++flags.DEFINE_string('kalign_binary_path', kalign_binary_path,
+                     'Path to the Kalign executable.')
+-flags.DEFINE_string('uniref90_database_path', None, 'Path to the Uniref90 '
++flags.DEFINE_string('uniref90_database_path', uniref90_database_path, 'Path to the Uniref90 '
+                     'database for use by JackHMMER.')
+-flags.DEFINE_string('mgnify_database_path', None, 'Path to the MGnify '
++flags.DEFINE_string('mgnify_database_path', mgnify_database_path, 'Path to the MGnify '
+                     'database for use by JackHMMER.')
+-flags.DEFINE_string('bfd_database_path', None, 'Path to the BFD '
++flags.DEFINE_string('bfd_database_path', bfd_database_path, 'Path to the BFD '
+                     'database for use by HHblits.')
+-flags.DEFINE_string('uniclust30_database_path', None, 'Path to the Uniclust30 '
++flags.DEFINE_string('uniclust30_database_path', uniclust30_database_path, 'Path to the Uniclust30 '
+                     'database for use by HHblits.')
+-flags.DEFINE_string('pdb70_database_path', None, 'Path to the PDB70 '
++flags.DEFINE_string('pdb70_database_path', pdb70_database_path, 'Path to the PDB70 '
+                     'database for use by HHsearch.')
+-flags.DEFINE_string('template_mmcif_dir', None, 'Path to a directory with '
++flags.DEFINE_string('template_mmcif_dir', template_mmcif_dir, 'Path to a directory with '
+                     'template mmCIF structures, each named <pdb_id>.cif')
+ flags.DEFINE_string('max_template_date', None, 'Maximum template release date '
+                     'to consider. Important if folding historical test sets.')
+-flags.DEFINE_string('obsolete_pdbs_path', None, 'Path to file containing a '
++flags.DEFINE_string('obsolete_pdbs_path', obsolete_pdbs_path, 'Path to file containing a '
+                     'mapping from obsolete PDB IDs to the PDB IDs of their '
+                     'replacements.')
+ flags.DEFINE_enum('preset', 'full_dbs', ['full_dbs', 'casp14'],
+@@ -268,16 +307,8 @@ if __name__ == '__main__':
+       'fasta_paths',
+       'output_dir',
+       'model_names',
+-      'data_dir',
+       'preset',
+-      'uniref90_database_path',
+-      'mgnify_database_path',
+-      'uniclust30_database_path',
+-      'bfd_database_path',
+-      'pdb70_database_path',
+-      'template_mmcif_dir',
+       'max_template_date',
+-      'obsolete_pdbs_path',
+   ])
+
+   app.run(main)

easybuild/easyconfigs/a/AlphaFold/AlphaFold-2.0.0_fix-packages.patch

@@ -0,0 +1,13 @@
+also include alphafold.data.tools and alphafold.model.tf, which are not picked up by find_packages()
+author: Kenneth Hoste (HPC-UGent)
+--- alphafold-2.0.0/setup.py.orig	2021-08-14 08:42:43.030919330 +0200
++++ alphafold-2.0.0/setup.py	2021-08-14 08:43:59.183245602 +0200
+@@ -26,7 +26,7 @@
+     author_email='alphafold@deepmind.com',
+     license='Apache License, Version 2.0',
+     url='https://github.com/deepmind/alphafold',
+-    packages=find_packages(),
++    packages=find_packages() + ['alphafold/data/tools', 'alphafold/model/tf'],
+     install_requires=[
+         'absl-py',
+         'biopython',

easybuild/easyconfigs/a/AlphaFold/AlphaFold-2.0.0_fix-scp-path.patch

@@ -0,0 +1,23 @@
+fix path to stereo_chemical_props.txt when AlphaFold is installed with EasyBuild
+author: Kenneth Hoste (HPC-UGent)
+diff --git a/alphafold/common/residue_constants.py b/alphafold/common/residue_constants.py
+index 318b44a..fd840e3 100644
+--- a/alphafold/common/residue_constants.py
++++ b/alphafold/common/residue_constants.py
+@@ -19,6 +19,7 @@ import functools
+ from typing import Mapping, List, Tuple
+
+ import numpy as np
++import os
+ import tree
+
+ # Internal import (35fd).
+@@ -403,7 +404,7 @@ def load_stereo_chemical_props() -> Tuple[Mapping[str, List[Bond]],
+     residue_bond_angles: dict that maps resname --> list of BondAngle tuples
+   """
+   stereo_chemical_props_path = (
+-      'alphafold/common/stereo_chemical_props.txt')
++      os.path.join(os.getenv('EBROOTALPHAFOLD', 'alphafold/common'), 'stereo_chemical_props.txt'))
+   with open(stereo_chemical_props_path, 'rt') as f:
+     stereo_chemical_props = f.read()
+   lines_iter = iter(stereo_chemical_props.splitlines())

easybuild/easyconfigs/a/AlphaFold/AlphaFold-2.0.0_n-cpu.patch

@@ -0,0 +1,49 @@
+don't hardcode number of cores to use by HHBlits and Jackhmmer scripts;
+see https://github.com/deepmind/alphafold/issues/110
+author: Kenneth Hoste (HPC-UGent)
+
+diff --git a/alphafold/data/tools/hhblits.py b/alphafold/data/tools/hhblits.py
+index e0aa098..e624f3a 100644
+--- a/alphafold/data/tools/hhblits.py
++++ b/alphafold/data/tools/hhblits.py
+@@ -27,6 +27,8 @@ from alphafold.data.tools import utils
+ _HHBLITS_DEFAULT_P = 20
+ _HHBLITS_DEFAULT_Z = 500
+
++HHBLITS_N_CPU = os.getenv('ALPHAFOLD_HHBLITS_N_CPU', 4)
++
+
+ class HHBlits:
+   """Python wrapper of the HHblits binary."""
+@@ -35,7 +37,7 @@ class HHBlits:
+                *,
+                binary_path: str,
+                databases: Sequence[str],
+-               n_cpu: int = 4,
++               n_cpu: int = HHBLITS_N_CPU,
+                n_iter: int = 3,
+                e_value: float = 0.001,
+                maxseq: int = 1_000_000,
+diff --git a/alphafold/data/tools/jackhmmer.py b/alphafold/data/tools/jackhmmer.py
+index ae53df9..800bf99 100644
+--- a/alphafold/data/tools/jackhmmer.py
++++ b/alphafold/data/tools/jackhmmer.py
+@@ -24,6 +24,9 @@ from alphafold.data.tools import utils
+ # Internal import (7716).
+
+
++JACKHMMER_N_CPU = os.getenv('ALPHAFOLD_JACKHMMER_N_CPU', 8)
++
++
+ class Jackhmmer:
+   """Python wrapper of the Jackhmmer binary."""
+
+@@ -31,7 +34,7 @@ class Jackhmmer:
+                *,
+                binary_path: str,
+                database_path: str,
+-               n_cpu: int = 8,
++               n_cpu: int = JACKHMMER_N_CPU,
+                n_iter: int = 1,
+                e_value: float = 0.0001,
+                z_value: Optional[int] = None,