Molecular dynamics

.github/workflows/main.yml

@@ -78,4 +78,64 @@ jobs:
       run: |  
         python -m pytest --nbval tests/PLM_prepare.py notebooks/PLM-interact.ipynb tests/PLM_output.py 
 
+  test_md:
+    runs-on: ubuntu-latest
+
+    strategy:
+      matrix:
+        include:
+          # Protein with small molecule with smiles
+          - input_file: md_inputs/G3_smol_model_0.cif
+            boltz_input_file: md_inputs/G3_smol_smi.yaml
+          # Protein ion as CCD
+          - input_file: md_inputs/G3_ion_model_0.cif
+            boltz_input_file: md_inputs/G3_ion.yaml
+          # Protein with RNA 
+          - input_file: md_inputs/G3_RNA_model_0.cif
+            boltz_input_file: md_inputs/G3_RNA.yaml
+
+    steps:
+      - uses: actions/checkout@v4
+
+      - uses: conda-incubator/setup-miniconda@fc2d68f6413eb2d87b895e92f8584b5b94a10167 # v3
+        with:
+          environment-file: environments/molecular_dynamics.yml
+          activate-environment: openmm
+          auto-activate-base: false
+
+      - name: Install local tool RNAprep
+        shell: bash -el {0}
+        run: |
+          pip install -e tools/RNAprep
+
+      - name: Install local tool ccd2smiles
+        shell: bash -el {0}
+        run: |
+          pip install -e tools/ccd2smiles
+
+      - name: Run ccd2smiles tests
+        run: |
+          source $CONDA/etc/profile.d/conda.sh
+          conda activate openmm
+          pytest tools/ccd2smiles/tests
+
+      - name: Patch notebook inputs
+        run: |
+          sed \
+            -e 's|__INPUT_FILE__|${{ matrix.input_file }}|g' \
+            -e 's|__BOLTZ_INPUT_FILE__|${{ matrix.boltz_input_file }}|g' \
+            notebooks/Molecular_Dynamics.ipynb > notebooks/Molecular_Dynamics_ci.ipynb
+
+      - name: Run MD Notebook
+        shell: bash -el {0}
+        run: |
+          python -m pytest --nbval tests/prepare_md.py notebooks/Molecular_Dynamics_ci.ipynb
+
+      - name: Run MD Analysis Notebook
+        if: matrix.input_file == 'md_inputs/G3_smol_model_0.cif'
+        shell: bash -el {0}
+        run: |
+          python -m pytest --nbval tests/prepare_md_output.py notebooks/MD_Analysis.ipynb 
+
+
 

docs/tutorials/tutorial_MD_bwVisu.md

@@ -0,0 +1,44 @@
+# Molecular Dynamics on bwVisu
+
+Welcome to the Molecular Dynamics Tutorial for bwVisu! 
+
+This tutorial will guide you through running short Molecular Dynamics Simulation on bwVisu. Please follow these steps carefully. Any feedback on the tutorial is welcome! Feel free to [contact us](../contact.md)!
+
+This tutorial is intended as a follow up to a structure prediction using either [AlphaFold](/docs/tutorials/tutorial_AF_bwVisu.md) or [Boltz](/docs/tutorials/tutorial_Boltz_bwVisu.md). 
+
+Note that only proteins, DNA and small molecules can be treated with this workflow. If you have more complex systems, [contact us](../contact.md) to discuss a solution.
+
+### Step 1: Access bwVisu and Start Jupyter
+
+Go to <a href="https://bwvisu.bwservices.uni-heidelberg.de/" target="_blank" rel="noopener">https://bwvisu.bwservices.uni-heidelberg.de/</a> and log in with your credentials and one-time password. 
+
+Choose Jupyter and start a new session. Now you can select the resources you need.
+
+For the Molecular Dynamics Simulation we will need a GPU of type A40.
+
+You also need access to the python environment, so set the `Kernel Path` to the OpenMM kernel at `/mnt/sds-hd/sd25g005/openmm/share/jupyter/`. 
+
+![Screenshot](../images/tutorial/bwVisu_GPU_Kernel.png){:.invertable}
+<!--{: style="height:500px;width:750px"}-->
+
+Click on "Launch". This will bring you to a new screen showing your interactive sessions. Wait for your session to be ready, then click on "Connect to Jupyter". This brings you into a JupyterLab environment.
+
+### Step 2: Go to your Working Directory and Upload Files
+
+Next all required files need to be present in the same directory as your predicted structure. Take the Molecular Dynamics notebook from our <a href="https://github.com/ssciwr/BioStructureHub/tree/main/notebooks" target="_blank" rel="noopener">github</a> and upload them by clicking on the upload button:
+
+![Screenshot](../images/tutorial/bwVisu_upload.png){: .invertable style="height:111px;width:444px"}
+
+Make sure that the output of your structure prediction is in the same directory as the `Molecular_Dynamics.ipynb` notebook. 
+
+- if you ran an **AlphaFold** prediction, you just need the `.cif` file
+- if you ran a **Boltz** prediction with small molecules present, you need the `.cif` file and the input `.yaml` file
+
+It should look like this:
+
+![Screenshot](../images/tutorial/bwVisu_MD_files.png){: .invertable style="height:142px"}
+
+
+### Step 3: Start the Simulation
+
+Open `Molecular_Dynamics.ipynb`, add your `.cif` file (and in case of Boltz your input `.yaml`) in the notebook and then execute all the cells in the notebook to start your Molecular Dynamics run!

environments/molecular_dynamics.yml

@@ -0,0 +1,23 @@
+name: openmm
+channels:
+  - conda-forge
+
+dependencies:
+  - python=3.12
+  - pip
+  - rdkit=2024.09.6
+  - openmm=8.4
+  - biopython
+  - nbval
+  - numpy=2.4.3
+  - pandas=2.3.3
+  - pdbfixer
+  - pytest
+  - scipy=1.17.1
+  - matplotlib=3.10.8
+  - mdanalysis
+  - openff-toolkit==0.18.0
+  - openmmforcefields
+
+
+

mkdocs.yml

@@ -59,6 +59,7 @@ nav:
     - tutorials/tutorial_Boltz_bwVisu.md
     - tutorials/tutorial_Boltzgen_bwVisu.md
     - tutorials/tutorial_RFDiffusion_bwVisu.md
+    - tutorials/tutorial_MD_bwVisu.md
     - tutorials/tutorial_use_templates.md
     - tutorials/tutorial_PLM_bwVisu.md
     - tutorials/tutorial_visu2helix.md

notebooks/MD_Visualization.ipynb

@@ -0,0 +1,140 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "5ab21147-78ae-414d-850e-06e6f5e05799",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# https://github.com/nglviewer/nglview\n",
+    "\n",
+    "# Uncomment below if you need to install nglview on the fly\n",
+    "# ! pip install jupyterlab_widgets ipywidgets nglview\n",
+    "# ! sed -i \"s/__frontend_version__ = '4.0'/__frontend_version__ = '3.1.5'/\" ~/.local/lib/python3.9/site-packages/nglview/_frontend.py"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "80cfd11b-a6a0-4b53-8e9b-c1a72b47e0f2",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "31d2bc2848324eff9dd3858a929b7477",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": []
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "import MDAnalysis as mda\n",
+    "import nglview as nv"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "7258bab2-5165-4430-817c-b4a61b89edad",
+   "metadata": {},
+   "source": [
+    "Insert your trajectory file:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "4a243c7a-f3a6-42c0-8332-0d076e69650e",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "input_trajectory = \"__TRAJECTORY_FILE__\""
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "8e5432a4-3fb0-448d-b43f-24b0454b2081",
+   "metadata": {},
+   "source": [
+    "Visualize the trajectory:"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "991a3b94-7361-4463-a553-432289622839",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "d7b2a533a88542689a91906aff8c026d",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "NGLWidget(max_frame=99)"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "u = mda.Universe(input_trajectory)\n",
+    "\n",
+    "view = nv.show_mdanalysis(u)\n",
+    "view.clear()\n",
+    "\n",
+    "# see more options here: https://nglviewer.org/ngl/api/manual/selection-language.html\n",
+    "view.add_representation(\"cartoon\", selection=\"protein\")\n",
+    "view.add_representation(\"licorice\", selection=\"water\", opacity=0.1)\n",
+    "view.center()\n",
+    "\n",
+    "view"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "fc92f7eb-0e38-40ae-ae71-af87887fde47",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "1b328850-f8e2-44cd-9819-fa5532697646",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.12.13"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}