![](\"oracle.png\")
"
]
},
{
"cell_type": "markdown",
"id": "compound-sapphire",
"metadata": {},
"source": [
"## Docking Oracle"
]
},
{
"cell_type": "markdown",
"id": "becoming-matrix",
"metadata": {},
"source": [
"* What is docking?\n",
"\n",
"Molecular docking is a computational approach to evaluate the binding affinity of a ligand (small molecule, flexible) and a receptor (protein target, rigid). Typically it includes two components: a scoring function to evaluate the free energy of a determined conformation, and a searching algorithm that search for the stronger binding conformation. A stronger binding affinity leads to higher probability to be a potent inhibiter (recall the lock and key model)."
]
},
{
"cell_type": "markdown",
"id": "rising-spice",
"metadata": {},
"source": [
"Bender, B. J., Gahbauer, S., Luttens, A., Lyu, J., Webb, C. M., Stein, R. M., ... & Shoichet, B. K. (2021). A practical guide to large-scale docking. Nature Protocols, 1-34."
]
},
{
"cell_type": "markdown",
"id": "continent-blink",
"metadata": {},
"source": [
"![](\"docking.png\")
"
]
},
{
"cell_type": "markdown",
"id": "multiple-refund",
"metadata": {},
"source": [
"* Why we use docking?\n",
"\n",
"1. Docking is actually used as a first step in virtual screening for drug discovery, more useful and challenging.\n",
"2. Docking serves as an affordable simulation of resource-consuming oracles (docking: ~30s/mol, QED: ms/mol)."
]
},
{
"cell_type": "markdown",
"id": "durable-representation",
"metadata": {},
"source": [
"![](\"why_docking.png\")
"
]
},
{
"cell_type": "markdown",
"id": "adapted-rocket",
"metadata": {},
"source": [
"## How to run docking (traditional vs. TDC) ?"
]
},
{
"cell_type": "markdown",
"id": "hazardous-fancy",
"metadata": {},
"source": [
"Traditional docking simulation: https://vina.scripps.edu/manual/#linux"
]
},
{
"cell_type": "markdown",
"id": "better-feeling",
"metadata": {},
"source": [
"![](\"vina.png\")
"
]
},
{
"cell_type": "markdown",
"id": "fuzzy-spare",
"metadata": {},
"source": [
"![](\"docking_gflownet.png\")
"
]
},
{
"cell_type": "markdown",
"id": "weekly-daisy",
"metadata": {},
"source": [
"We provide easy access with AutoDock Vina as backend via PyScreener, install PyScreener following https://github.com/coleygroup/pyscreener"
]
},
{
"cell_type": "markdown",
"id": "raising-median",
"metadata": {},
"source": [
"We provide 9 curated targets: PDB ID: 3pbl, 1iep, 2rgp, 3eml, 3ny8, 4rlu, 4unn, 5mo4, 7l11. Using them is similar to using other oracles:"
]
},
{
"cell_type": "markdown",
"id": "conditional-average",
"metadata": {},
"source": [
"| PDB ID | Description | Source |\n",
"| ----------- | ----------- | ----------- |\n",
"| [1iep](https://www.rcsb.org/structure/1IEP) | c-Abl kinase, complex with imatinib. | [vina tutorial](https://autodock-vina.readthedocs.io/en/latest/docking_python.html) |\n",
"| [2rgp](https://www.rcsb.org/structure/2RGP) | Epidermal growth factor receptor, complex with hydrazone. | [SBMolGen](https://chemrxiv.org/engage/chemrxiv/article-details/60c75725842e65c7acdb4638) |\n",
"| [3eml](https://www.rcsb.org/structure/3EML) | Human A2A adenosine receptor, complex with ZM241385. | [SBMolGen](https://chemrxiv.org/engage/chemrxiv/article-details/60c75725842e65c7acdb4638) |\n",
"| [3ny8](https://www.rcsb.org/structure/3NY8) | Beta2 adrenergic receptor, complex with the inverse agonist ICI 118,551. | [SBMolGen](https://chemrxiv.org/engage/chemrxiv/article-details/60c75725842e65c7acdb4638) |\n",
"| [3pbl](https://www.rcsb.org/structure/3PBL) | Human dopamine D3 receptor, complex with eticlopride. | [DUD.E](http://dude.docking.org/targets/drd3) |\n",
"| [4rlu](https://www.rcsb.org/structure/4RLU) | beta-hydroxyacyl-ACP dehydratase HadAB, complex with a flavonoid inhibitor. | [A 3D generative model for structure-based drug design](https://proceedings.neurips.cc/paper/2021/hash/314450613369e0ee72d0da7f6fee773c-Abstract.html) |\n",
"| [4unn](https://www.rcsb.org/structure/4UNN) | M. Tuberculosis Thymidylate Kinase (Mtb Tmk), complex witha. inhibitor. | [MolPAL](https://pubs.rsc.org/en/content/articlehtml/2021/sc/d0sc06805e) |\n",
"| [5mo4](https://www.rcsb.org/structure/5MO4) | ABL1 kinase, complex with asciminib. | [L-Net](https://arxiv.org/abs/2104.08474) |\n",
"| [7l11](https://www.rcsb.org/structure/7L11) | The Main Protease (M$^{pro}$) of SARS-CoV-2, complex with a known inhibitor. | [Zhang, Chun-Hui, et al.](https://pubs.acs.org/doi/abs/10.1021/acscentsci.1c00039) |"
]
},
{
"cell_type": "code",
"execution_count": 1,
"id": "comic-thesis",
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"Found local copy...\n",
"2022-01-24 23:17:41,348\tINFO services.py:1253 -- View the Ray dashboard at \u001b[1m\u001b[32mhttp://127.0.0.1:8266\u001b[39m\u001b[22m\n",
"Docking: 100%|██████████| 1/1 [00:23<00:00, 23.08s/ligand]\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Docking score: -8.0 kcal/mol\n",
"Consumed time: 23.086 s\n"
]
}
],
"source": [
"from tdc import Oracle\n",
"import time\n",
"\n",
"oracle = Oracle(name='3pbl_docking')\n",
"t1 = time.time()\n",
"print(f\"Docking score: {oracle('CCNC(=O)c1ccc(NC(=O)N2CC[C@H](C)[C@H](O)C2)c(C)c1')} kcal/mol\")\n",
"print(f\"Consumed time: {time.time() - t1:.3f} s\")"
]
},
{
"cell_type": "markdown",
"id": "proprietary-marketplace",
"metadata": {},
"source": [
"## Visualize the result"
]
},
{
"cell_type": "code",
"execution_count": 2,
"id": "quick-bennett",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"0"
]
},
"execution_count": 2,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"import py3Dmol\n",
"import os\n",
"\n",
"pdbid = '3pbl'\n",
"\n",
"# conda install openbabel -c conda-forge\n",
"os.system(\"obabel -i pdbqt out.pdbqt -o pdb -O \" + pdbid + \"_ligand.pdb\")\n",
"os.system(\"obabel -i pdbqt oracle/\" + pdbid + \".pdbqt -o pdb -O\" + pdbid + \"_receptor.pdb\")\n",
"os.system(\"obabel -i pdbqt out.pdbqt -o xyz -O \" + pdbid + \"_ligand.xyz\")"
]
},
{
"cell_type": "code",
"execution_count": 3,
"id": "relative-authorization",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"7.394807692307693 22.16196153846154 23.601230769230767\n"
]
},
{
"data": {
"application/3dmoljs_load.v0": "You appear to be running in JupyterLab (or JavaScript failed to load for some other reason). You need to install the 3dmol extension:
\n jupyter labextension install jupyterlab_3dmol
You appear to be running in JupyterLab (or JavaScript failed to load for some other reason). You need to install the 3dmol extension:
\n",
" jupyter labextension install jupyterlab_3dmol
| \n", " | smiles | \n", "
|---|---|
| 0 | \n", "CC(C)(C)c1ccc2occ(CC(=O)Nc3ccccc3F)c2c1 | \n", "
| 1 | \n", "C[C@@H]1CC(Nc2cncc(-c3nncn3C)c2)C[C@@H](C)C1 | \n", "
| 2 | \n", "N#Cc1ccc(-c2ccc(O[C@@H](C(=O)N3CCCC3)c3ccccc3)... | \n", "
| 3 | \n", "CCOC(=O)[C@@H]1CCCN(C(=O)c2nc(-c3ccc(C)cc3)n3c... | \n", "
| 4 | \n", "N#CC1=C(SCC(=O)Nc2cccc(Cl)c2)N=C([O-])[C@H](C#... | \n", "
![](\"generation_process.png\")
"
]
},
{
"cell_type": "markdown",
"id": "adopted-tutorial",
"metadata": {},
"source": [
"