Merge pull request #122 from adaptyvbio/extra_dependency

Make some dependencies optional
adaptyvbio · Nov 15, 2023 · b0a6ca5 · b0a6ca5
2 parents 8d661bd + e86ed1c
commit b0a6ca5
Show file tree

Hide file tree

Showing 11 changed files with 244 additions and 32 deletions.
diff --git a/.conda/arm64/meta.yaml b/.conda/arm64/meta.yaml
@@ -25,12 +25,12 @@ requirements:
     - wheel
     - setuptools
   run:
-    - numpy>=1.17
+    - numpy>=1.21,<1.25
     - editdistance>=0.6.0
     - Biopython==1.79 
     - click>=8.1.3
     - biopandas>=0.4.1
-    - boto3==1.24.59
+    - boto3
     - p-tqdm
     - networkx==2.8.8
     - einops
@@ -41,8 +41,10 @@ requirements:
     - awscli==1.25.60
     - bs4
     - rcsbsearch
-    - tmtools
-    - fair-esm
+    - pyyaml
+    - rdkit
+    - pypdb
+    - joblib
 
 about:
   home: https://github.com/adaptyvbio/ProteinFlow

diff --git a/.conda/default/meta.yaml b/.conda/default/meta.yaml
@@ -25,26 +25,28 @@ requirements:
     - wheel
     - setuptools
   run:
-    - numpy>=1.17
+    - numpy>=1.21,<1.25
     - editdistance>=0.6.0
     - Biopython==1.79 
     - click>=8.1.3
     - biopandas>=0.4.1
-    - boto3==1.24.59
+    - boto3=
     - p-tqdm
     - networkx==2.8.8
     - einops
     - pandas
     - pytorch>=1.10.0
     - biotite==0.35.0
-    - aiobotocore==2.4.2
-    - awscli==1.25.60
+    - aiobotocore
+    - awscli
     - bs4
     - rcsbsearch
-    - tmtools
-    - fair-esm
     - mmseqs2
     - foldseek
+    - pyyaml
+    - rdkit
+    - pypdb
+    - joblib
 
 about:
   home: https://github.com/adaptyvbio/ProteinFlow

diff --git a/channeldata.json b/channeldata.json
@@ -0,0 +1,39 @@
+{
+  "channeldata_version": 1,
+  "packages": {
+    "proteinflow": {
+      "activate.d": false,
+      "binary_prefix": false,
+      "deactivate.d": false,
+      "description": null,
+      "dev_url": "https://github.com/adaptyvbio/ProteinFlow",
+      "doc_source_url": null,
+      "doc_url": "https://adaptyvbio.github.io/ProteinFlow/",
+      "home": "https://github.com/adaptyvbio/ProteinFlow",
+      "icon_hash": null,
+      "icon_url": null,
+      "identifiers": null,
+      "keywords": null,
+      "license": "BSD-3-Clause",
+      "post_link": false,
+      "pre_link": false,
+      "pre_unlink": false,
+      "recipe_origin": null,
+      "run_exports": {},
+      "source_git_url": "https://github.com/adaptyvbio/ProteinFlow.git",
+      "source_url": null,
+      "subdirs": [
+        "linux-64"
+      ],
+      "summary": "Versatile pipeline for processing protein structure data for deep learning applications.",
+      "tags": null,
+      "text_prefix": true,
+      "timestamp": 1700064405,
+      "version": "2.4.1"
+    }
+  },
+  "subdirs": [
+    "linux-64",
+    "noarch"
+  ]
+}
diff --git a/index.html b/index.html
@@ -0,0 +1,90 @@
+<html>
+<head>
+  <title>proteinflow</title>
+  <style type="text/css">
+    a, a:active {
+      text-decoration: none; color: blue;
+    }
+    a:visited {
+      color: #48468F;
+    }
+    a:hover, a:focus {
+      text-decoration: underline; color: red;
+    }
+    body {
+      background-color: #F5F5F5;
+    }
+    h2 {
+      margin-bottom: 12px;
+    }
+    th, td {
+      font: 100% monospace; text-align: left;
+    }
+    th {
+      font-weight: bold; padding-right: 14px; padding-bottom: 3px;
+    }
+    th.tight {
+        padding-right: 6px;
+    }
+    td {
+      padding-right: 14px;
+    }
+    td.tight {
+        padding-right: 8px;
+    }
+    td.s, th.s {
+      text-align: right;
+    }
+    td.summary {
+      white-space: nowrap;
+      overflow: hidden;
+    }
+    td.packagename {
+      white-space: nowrap;
+      text-overflow: ellipsis;
+      overflow: hidden;
+      max-width: 180px;
+      padding-right: 8px;
+    }
+    td.version {
+      //white-space: nowrap;
+      overflow: hidden;
+      max-width: 90px;
+      padding-right: 8px;
+    }
+    table {
+      background-color: white;
+      border-top: 1px solid #646464;
+      border-bottom: 1px solid #646464;
+      padding-top: 10px;
+      padding-bottom: 14px;
+    }
+    address {
+      color: #787878;
+      padding-top: 10px;
+    }
+  </style>
+</head>
+<body>
+  <h2>proteinflow</h2>
+  <h3><a href="rss.xml">RSS Feed</a>&nbsp;&nbsp;&nbsp;<a href="channeldata.json">channeldata.json</a></h3>
+<a href="linux-64">linux-64</a>&nbsp;&nbsp;&nbsp;<a href="noarch">noarch</a>&nbsp;&nbsp;&nbsp;  <table>
+    <tr>
+      <th style="padding-right:18px;">Package</th>
+      <th>Latest Version</th>
+      <th>Doc</th>
+      <th>Dev</th>
+      <th>License</th>
+<th class="tight">linux-64</th><th class="tight">noarch</th>      <th>Summary</th>
+    </tr>
+    <tr>
+      <td class="packagename"><a href="https://github.com/adaptyvbio/ProteinFlow" alt="proteinflow">proteinflow</a></td>
+      <td class="version">2.4.1</td>
+      <td><a href="https://adaptyvbio.github.io/ProteinFlow/">doc</a></td>
+      <td><a href="https://github.com/adaptyvbio/ProteinFlow">dev</a></td>
+      <td class="tight">BSD-3-Clause</td>
+<td>X</td><td></td>      <td class="summary">Versatile pipeline for processing protein structure data for deep...</td>
+    </tr>  </table>
+  <address>Updated: 2023-11-15 16:08:06 +0000 - Files: 1</address>
+</body>
+</html>
diff --git a/install_optional.sh b/install_optional.sh
@@ -8,7 +8,5 @@ python -m pip install "fair-esm[esmfold]"
 python -m pip install 'dllogger @ git+https://github.com/NVIDIA/dllogger.git'
 python -m pip install 'openfold @ git+https://github.com/aqlaboratory/openfold.git@4b41059694619831a7db195b7e0988fc4ff3a307'
 
-python -m pip install ablang igfold immunebuilder
-
 python -m pip install -e .
-python -m pip install ipykernel
+# python -m pip install ipykernel
diff --git a/proteinflow/__init__.py b/proteinflow/__init__.py
@@ -171,6 +171,7 @@
     "split": False,
     "cli": False,
     "ligand": False,
+    "extra": False,
 }
 __docformat__ = "numpy"
 

diff --git a/proteinflow/data/__init__.py b/proteinflow/data/__init__.py
@@ -18,18 +18,31 @@
 from collections import defaultdict
 
 import Bio.PDB
-import MDAnalysis as mda
 import numpy as np
 import pandas as pd
-import py3Dmol
 from Bio import pairwise2
 from biopandas.pdb import PandasPdb
-from methodtools import lru_cache
 from torch import Tensor, from_numpy
 
+try:
+    import MDAnalysis as mda
+except ImportError:
+    pass
+try:
+    from methodtools import lru_cache
+except ImportError:
+
+    def lru_cache():
+        """Make a dummy decorator."""
+
+        def wrapper(func):
+            return func
+
+        return wrapper
+
+
 from proteinflow.constants import (
     _PMAP,
-    ACCENT_COLOR,
     ALPHABET,
     ALPHABET_REVERSE,
     ATOM_MASKS,
@@ -52,6 +65,7 @@
     _retrieve_chain_names,
 )
 from proteinflow.download import download_fasta, download_pdb
+from proteinflow.extra import _get_view, requires_extra
 from proteinflow.ligand import _get_ligands
 from proteinflow.metrics import (
     ablang_pll,
@@ -1979,6 +1993,7 @@ def align_structure(self, reference_pdb_path, save_pdb_path, chain_ids=None):
         io.save(save_pdb_path)
 
     @staticmethod
+    @requires_extra("MDAnalysis")
     def combine_multiple_frames(files, output_path="combined.pdb"):
         """Combine multiple PDB files into a single multiframe PDB file.
 
@@ -2570,7 +2585,7 @@ def visualize(
             accent_color=accent_color,
         )
         vis_string = "".join([str(x) for x in outstr])
-        view = py3Dmol.view(width=canvas_size[0], height=canvas_size[1])
+        view = _get_view(canvas_size)
         view.addModelsAsFrames(vis_string)
         for i, at in enumerate(outstr):
             view.setStyle(

diff --git a/proteinflow/extra.py b/proteinflow/extra.py
@@ -0,0 +1,41 @@
+"""Handling optional dependencies."""
+
+try:
+    import py3Dmol
+except ImportError:
+    pass
+
+import sys
+
+
+def requires_extra(module_name, install_name=None):
+    """Generate a decorator to require an optional dependency for the given function.
+
+    Parameters
+    ----------
+    module_name : str
+        Name of the module to check for
+    install_name : str, optional
+        Name of the module to install if it is not found. If not specified, `module_name` is used
+
+    """
+    if install_name is None:
+        install_name = module_name
+
+    def decorator(func):
+        def wrapper(*args, **kwargs):
+            if module_name not in sys.modules:
+                raise ImportError(
+                    f"{install_name} must be installed to use this function. "
+                    f"Install it with `pip install {install_name}` or together with most other optional dependencies with `pip install proteinflow[processing]`."
+                )
+            return func(*args, **kwargs)
+
+        return wrapper
+
+    return decorator
+
+
+@requires_extra("py3Dmol")
+def _get_view(canvas_size):
+    return py3Dmol.view(width=canvas_size[0], height=canvas_size[1])