Merge pull request #368 from volkamerlab/dev

[2023.05.2-base] DL_edition release prep
volkamerlab · Jun 2, 2023 · d875d36 · d875d36
2 parents b2b75e1 + 54125aa
commit d875d36
Show file tree

Hide file tree

Showing 96 changed files with 68,216 additions and 3,054 deletions.
diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml
@@ -73,18 +73,33 @@ jobs:
         run: |
           PYTEST_ARGS="--nbval-lax --current-env --dist loadscope --numprocesses 2"
 
+          PYTEST_IGNORE_T001="--ignore=teachopencadd/talktorials/T001_query_chembl/talktorial.ipynb"
+          PYTEST_IGNORE_T018="--ignore=teachopencadd/talktorials/T018_automated_cadd_pipeline/talktorial.ipynb"
           PYTEST_IGNORE_T019="--ignore=teachopencadd/talktorials/T019_md_simulation/talktorial.ipynb"
           PYTEST_IGNORE_T020="--ignore=teachopencadd/talktorials/T020_md_analysis/talktorial.ipynb"
+          PYTEST_IGNORE_T035="--ignore=teachopencadd/talktorials/T035_graph_neural_networks/talktorial.ipynb"
+          PYTEST_IGNORE_T036="--ignore=teachopencadd/talktorials/T036_e3_equivariant_gnn/talktorial.ipynb"
+          PYTEST_IGNORE_T038="--ignore=teachopencadd/talktorials/T038_protein_ligand_interaction_prediction/talktorial.ipynb"
 
-          IGNORE=""
+          IGNORE="$PYTEST_IGNORE_T001"
 
           if [ "$RUNNER_OS" == "Windows" ]; then
             # Ignore T019 under Windows, see https://github.com/volkamerlab/teachopencadd/issues/313
             IGNORE="$IGNORE $PYTEST_IGNORE_T019"
+            IGNORE="$IGNORE $PYTEST_IGNORE_T035"
+            IGNORE="$IGNORE $PYTEST_IGNORE_T036"
+            IGNORE="$IGNORE $PYTEST_IGNORE_T038"
           fi
-
+          
+          if [ "$RUNNER_OS" == "Linux" ]; then
+            IGNORE="$IGNORE $PYTEST_IGNORE_T018" # see 385
+          fi
+          
           if [ "$RUNNER_OS" == "macOS" ]; then
+            IGNORE="$IGNORE $PYTEST_IGNORE_T018" # see 385
             IGNORE="$IGNORE $PYTEST_IGNORE_T019"
+            IGNORE="$IGNORE $PYTEST_IGNORE_T035" # may cause kernel dying in pytest
+            IGNORE="$IGNORE $PYTEST_IGNORE_T036" # may cause kernel dying in pytest
           fi
 
           pytest $PYTEST_ARGS teachopencadd/talktorials/ $IGNORE

diff --git a/.github/workflows/docs.yml b/.github/workflows/docs.yml
@@ -8,6 +8,7 @@ on:
   pull_request:
     branches:
       - "master"
+      - "dev"
       - "maintenance/.+"
   schedule:
     # Run a cron job once weekly on Monday
@@ -57,27 +58,28 @@ jobs:
           sed -i 's|src="images/butina_full.pdf"|src="../_images/butina_full.pdf"|g' \
             _build/html/talktorials/T005_compound_clustering.html
 
-      - name: Check 404s (Sphinx)
-        shell: bash -l {0}
-        run: |
-          cd docs/_build/html
-          npx linkinator . --verbosity error --recurse -r --retry-errors-jitter --timeout 10000 \
-            -s "/*.txt" \
-            -s "https://onlinelibrary.wiley.com/doi/full/10.1002/cmdc.200700139" \
-            -s "https://onlinelibrary.wiley.com/doi/10.1002/9783527806539.*" \
-            -s "https://towardsdatascience.com" \
-            -s "https://medium.com" \
-            -s "https://*.medium.com" \
-            -s "http[s]://quora.com" \
-            -s "http[s]://doi.org/*" \
-            -s "http[s]://pubs.acs.org/*" \
-            -s "ftp://*" \
-            -s "https://machinelearningmastery.com/*" \
-            -s "https://www.schrodinger.com/training/videos/*" \
-            -s "/*.mp4" \
-            -s "/*.pdf" \
-            -s "/*.png" \
-            -s "https://www.icoa.fr/pkidb/index.html" \
+      #- name: Check 404s (Sphinx)
+        #shell: bash -l {0}
+        #run: |
+          #cd docs/_build/html
+          #npx linkinator . --verbosity error --recurse -r --retry-errors-jitter --timeout 10000 \
+            #-s "/*.txt" \
+            #-s "https://onlinelibrary.wiley.com/doi/*" \
+            #-s "https://wires.onlinelibrary.wiley.com/doi/*" \
+            #-s "https://chemistry-europe.onlinelibrary.wiley.com/doi/*" \
+            #-s "https://towardsdatascience.com" \
+            #-s "https://medium.com" \
+            #-s "https://*.medium.com" \
+            #-s "http[s]://quora.com" \
+            #-s "http[s]://doi.org/*" \
+            #-s "http[s]://pubs.acs.org/*" \
+            #-s "ftp://*" \
+            #-s "https://machinelearningmastery.com/*" \
+            #-s "https://www.schrodinger.com/training/videos/*" \
+            #-s "/*.mp4" \
+            #-s "/*.pdf" \
+            #-s "/*.png" \
+            #-s "https://www.icoa.fr/pkidb/index.html" \
 
 
       - name: Check 404s (README)

diff --git a/.gitignore b/.gitignore
@@ -155,4 +155,10 @@ node_modules/
 # Talktorial outputs
 # T018
 teachopencadd/talktorials/T018_automated_cadd_pipeline/data/Outputs
-teachopencadd/talktorials/T018_automated_cadd_pipeline/data/PipelineInputData_Project2.csv
+teachopencadd/talktorials/T018_automated_cadd_pipeline/data/PipelineInputData_Project2.csv
+
+# Taltorial TO037 QM9 too large temp. fix
+teachopencadd/talktorials/T037_e3_equivariant_gnn/data/raw
+teachopencadd/talktorials/T037_e3_equivariant_gnn/data/processed
+
+
diff --git a/devtools/test_env.yml b/devtools/test_env.yml
@@ -3,14 +3,18 @@ channels:
   - conda-forge
   - defaults
 dependencies:
+  - mkl < 2022
+  - pytorch::pytorch==1.13.0
+  #- pyg::pyg==2.2.0
   - python>=3.8
   - pip
   - jupyterlab>=3
   # Workaround for jupyterlab, see https://github.com/volkamerlab/teachopencadd/issues/310
   - jsonschema>=4.3.0
   - nglview>=3
   # Workaround for nglview, see https://github.com/volkamerlab/teachopencadd/issues/262
-  - ipywidgets<8
+  - ipywidgets<=7.75
+  - jupyter_client<=7.4.9
   # New numpy version 1.2.4 too young, e.g. caused
   # https://github.com/volkamerlab/teachopencadd/issues/299
   - numpy<1.24
@@ -65,5 +69,6 @@ dependencies:
       - black-nb
       - nbsphinx-link
       - sphinxext-opengraph
+      - deepchem
       # TeachOpenCADD itself
       - ../
diff --git a/docs/_static/images/TeachOpenCADD_topics.png b/docs/_static/images/TeachOpenCADD_topics.png
diff --git a/docs/all_talktorials.rst b/docs/all_talktorials.rst
@@ -34,3 +34,9 @@ This is the complete list of talktorials available for online reading. Take into
    talktorials/T026_kinase_similarity_ifp.nblink
    talktorials/T027_kinase_similarity_ligand_profile.nblink
    talktorials/T028_kinase_similarity_compare_perspectives.nblink
+   talktorials/T033_molecular_representations.nblink
+   talktorials/T034_recurrent_neural_networks.nblink
+   talktorials/T035_graph_neural_networks.nblink
+   talktorials/T036_e3_equivariant_gnn.nblink
+   talktorials/T037_uncertainty_estimation.nblink
+   talktorials/T038_protein_ligand_interaction_prediction.nblink
diff --git a/docs/citation.rst b/docs/citation.rst
@@ -45,6 +45,17 @@ TeachOpenCADD Jupyter notebooks on kinase similarities spanning `talktorials T02
         doi = {10.1186/s13321-019-0351-x},
     }
 
+TeachOpenCADD Jupyter notebooks on deep learning spanning `talktorials T033-T038 (paper) <https://doi.org/10.26434/chemrxiv-2023-kz1pb>`_::
+
+   @article{TeachOpenCADD2023,
+        author = {Backenköhler, Michael and Kramer, Paula Linh and Groß, Joschka and Großmann, Gerrit and Joeres, Roman and Tagirdzhanov, Azat and Sydow, Dominique and Ibrahim, Hamza and Odje, Floriane and Wolf, Verena and Volkamer, Andrea},
+        title = {{TeachOpenCADD goes Deep Learning: Open-source Teaching Platform Exploring Molecular DL Applications}},
+        journal = {ChemRxiv preprint},
+        year = {2023}, 
+        doi = {10.26434/chemrxiv-2023-kz1pb},
+   } 
+
+
 TeachOpenCADD-KNIME
 -------------------
 
@@ -73,4 +84,4 @@ How to use the TeachOpenCADD material for `teaching (book chapter) <https://pubs
         chapter = {10},
         pages = {135-158},
         doi = {10.1021/bk-2021-1387.ch010},
-    }
+    }
diff --git a/docs/external_dependencies.rst b/docs/external_dependencies.rst
@@ -24,6 +24,11 @@ Python packages
   - ``biotite``: https://www.biotite-python.org/
   - ``smina``: https://sourceforge.net/p/smina/discussion/
 
+- Deep learning: 
+
+  - ``pytorch``: https://pytorch.org/
+  - ``pyg``: https://pyg.org/
+
 - Data science (PyData stack):
 
   - ``numpy``: https://numpy.org/
@@ -78,4 +83,4 @@ Databases and webservers
 - RCSB PDB: https://www.rcsb.org/
 - KLIFS: https://klifs.net/
 - PubMed: https://pubchem.ncbi.nlm.nih.gov/
-- ProteinsPlus: https://proteins.plus/
+- ProteinsPlus: https://proteins.plus/
diff --git a/docs/talktorials.rst b/docs/talktorials.rst
@@ -115,3 +115,18 @@ Kinase similarity comparison using various methods
    talktorials/T027_kinase_similarity_ligand_profile.nblink
    talktorials/T028_kinase_similarity_compare_perspectives.nblink
 
+
+Deep learning 
+-------------------
+
+Various deep learning architectures for drug discovery tasks
+
+.. nbgallery::
+   :name: deeplearning
+
+   talktorials/T033_molecular_representations.nblink
+   talktorials/T034_recurrent_neural_networks.nblink
+   talktorials/T035_graph_neural_networks.nblink
+   talktorials/T036_e3_equivariant_gnn.nblink
+   talktorials/T037_uncertainty_estimation.nblink
+   talktorials/T038_protein_ligand_interaction_prediction.nblink
diff --git a/docs/talktorials/T033_molecular_representations.nblink b/docs/talktorials/T033_molecular_representations.nblink
@@ -0,0 +1 @@
+{"path": "../../teachopencadd/talktorials/T033_molecular_representations/talktorial.ipynb", "extra-media": ["../../teachopencadd/talktorials/T033_molecular_representations/images"]}
diff --git a/docs/talktorials/T034_recurrent_neural_networks.nblink b/docs/talktorials/T034_recurrent_neural_networks.nblink
@@ -0,0 +1 @@
+{"path": "../../teachopencadd/talktorials/T034_recurrent_neural_networks/talktorial.ipynb", "extra-media": ["../../teachopencadd/talktorials/T034_recurrent_neural_networks/images"]}
diff --git a/docs/talktorials/T035_graph_neural_networks.nblink b/docs/talktorials/T035_graph_neural_networks.nblink
@@ -0,0 +1 @@
+{"path": "../../teachopencadd/talktorials/T035_graph_neural_networks/talktorial.ipynb", "extra-media": ["../../teachopencadd/talktorials/T035_graph_neural_networks/images"]}
diff --git a/docs/talktorials/T036_e3_equivariant_gnn.nblink b/docs/talktorials/T036_e3_equivariant_gnn.nblink
@@ -0,0 +1 @@
+{"path": "../../teachopencadd/talktorials/T036_e3_equivariant_gnn/talktorial.ipynb", "extra-media": ["../../teachopencadd/talktorials/T036_e3_equivariant_gnn/images"]}
diff --git a/docs/talktorials/T037_uncertainty_estimation.nblink b/docs/talktorials/T037_uncertainty_estimation.nblink
@@ -0,0 +1 @@
+{"path": "../../teachopencadd/talktorials/T037_uncertainty_estimation/talktorial.ipynb", "extra-media": ["../../teachopencadd/talktorials/T037_uncertainty_estimation/images"]}
diff --git a/docs/talktorials/T038_protein_ligand_interaction_prediction.nblink b/docs/talktorials/T038_protein_ligand_interaction_prediction.nblink
@@ -0,0 +1 @@
+{"path": "../../teachopencadd/talktorials/T038_protein_ligand_interaction_prediction/talktorial.ipynb", "extra-media": ["../../teachopencadd/talktorials/T038_protein_ligand_interaction_prediction/images"]}
diff --git a/teachopencadd/talktorials/T001_query_chembl/README.md b/teachopencadd/talktorials/T001_query_chembl/README.md
@@ -18,14 +18,14 @@ __Talktorial T001__: This talktorial is part of the TeachOpenCADD pipeline descr
 
 In this notebook, we will learn more about the ChEMBL database and how to extract data from ChEMBL, i.e. (compound, activity data) pairs for a target of interest. These data sets can be used for many cheminformatics tasks, such as similarity search, clustering or machine learning.
 
-Our work here will include finding compounds which were tested against a certain target and filtering available bioactivity data.
+Our work here will include finding compounds that were tested against a certain target and filtering available bioactivity data.
 
 
 ### Contents in *Theory*
 
 * ChEMBL database
     * ChEMBL web services
-    * ChEMBL webresource client
+    * ChEMBL web resource client
 * Compound activity measures
     * IC50 measure
     * pIC50 value
@@ -40,7 +40,7 @@ Our work here will include finding compounds which were tested against a certain
     * Fetch and download target data
     * Select target ChEMBL ID
 * Get bioactivity data
-    * Fetch and download bioactivity data for target
+    * Fetch and download bioactivity data for targets
     * Preprocess and filter bioactivity data
 * Get compound data
     * Fetch and download compound data
@@ -57,7 +57,7 @@ Our work here will include finding compounds which were tested against a certain
 * ChEMBL bioactivity database: [Gaulton *et al.*, <i>Nucleic Acids Res.</i> (2017), 45(Database issue), D945–D954](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5210557/)
 * ChEMBL web services: [Davies *et al.*, <i>Nucleic Acids Res.</i> (2015), <b>43</b>, 612-620](https://academic.oup.com/nar/article/43/W1/W612/2467881) 
 * [ChEMBL web-interface](https://www.ebi.ac.uk/chembl/)
-*  GitHub [ChEMBL webrescource client](https://github.com/chembl/chembl_webresource_client)
+*  GitHub [ChEMBL web rescource client](https://github.com/chembl/chembl_webresource_client)
 * The EBI RDF platform: [Jupp *et al.*, <i>Bioinformatics </i> (2014), 30(9), 1338-9](https://www.ncbi.nlm.nih.gov/pubmed/24413672)
 * Info on half maximal inhibitory concentration: [(p)IC50](https://en.wikipedia.org/wiki/IC50)
 * [UniProt website](https://www.uniprot.org/)
Original file line number	Diff line number	Diff line change
		@@ -0,0 +1 @@
		{"path": "../../teachopencadd/talktorials/T033_molecular_representations/talktorial.ipynb", "extra-media": ["../../teachopencadd/talktorials/T033_molecular_representations/images"]}