Testing cset-uload

test-cset.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "b00a8c4f-11fa-4127-84ee-d57e2a6982ca",
   "metadata": {},
   "outputs": [],
   "source": [
    "from rdkit import Chem, rdBase\n",
    "from rdkit.Chem import AllChem, PandasTools, Draw\n",
    "from rdkit.Chem.Draw import IPythonConsole\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "3f2dada0-8d63-4718-86cb-81538b03d2b8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "20638\n",
      "100\n",
      "100\n"
     ]
    }
   ],
   "source": [
    "# Wikipedia compounds view counts as a test set\n",
    "\n",
    "compounds = pd.read_csv('https://raw.githubusercontent.com/matteoferla/Wikipedian-compounds/main/compounds.csv').rename(columns=dict(combined_SMILES='SMILES'))\n",
    "print(len(compounds))\n",
    "compounds = compounds.loc[(~compounds.SMILES.isna()) & (compounds.SMILES != '}}')].head(100).copy()\n",
    "print(len(compounds))\n",
    "with rdBase.BlockLogs() as blocker:\n",
    "    PandasTools.AddMoleculeColumnToFrame(compounds, 'SMILES', 'mol')\n",
    "    compounds = compounds.loc[compounds.mol.astype(bool)].copy()\n",
    "    compounds.mol.apply(AllChem.EmbedMolecule)\n",
    "print(len(compounds))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "8e148cd7-c31e-44d8-b8e9-1fb1a43af82e",
   "metadata": {},
   "outputs": [],
   "source": [
    "from gist_import import GistImporter\n",
    "\n",
    "# fu for fragalysis upload\n",
    "fmodule = GistImporter.from_github('https://raw.githubusercontent.com/matteoferla/Fragment-hit-follow-up-chemistry/main/followup/prep_fragalysis.py')\n",
    "prep = fmodule['prep']\n",
    "generate_header = fmodule['generate_header']"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "69a07ce7-8ce9-4fc9-9962-b21fb26da93c",
   "metadata": {},
   "source": [
    "# Test 1: NUDT7A\n",
    "\n",
    "Killian's NUDT7A chosen as testcase."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "88e212ba-d3d9-4c13-af49-cc758a9fe6d4",
   "metadata": {},
   "outputs": [],
   "source": [
    "compounds = compounds.loc[compounds.SMILES.apply(len) < 200].copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "18e56cc1-7f4b-4562-92af-8d6aaff19f79",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['title', 'is_element', 'monthly_views', 'SMILES', 'mol'], dtype='object')"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "compounds.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "0745ef5e-ce4f-4d81-b7c7-073cc2412a9c",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[12:22:37] Molecule does not have explicit Hs. Consider calling AddHs()\n"
     ]
    }
   ],
   "source": [
    "test_name = 'MF-test231113_base'\n",
    "\n",
    "header: Chem.Mol = generate_header(method=test_name,\n",
    "                         ref_url='https://www.example.com',\n",
    "                         submitter_name='matteo',\n",
    "                         submitter_email='matteo.ferla@stats.ox.ac.uk',\n",
    "                         submitter_institution='Ox',\n",
    "                         extras={'monthly_views': 'monthly views in Wikipedia'}\n",
    "                                  )\n",
    "\n",
    "prep(compounds, \n",
    "     header, mol_col='mol', \n",
    "     name_col='title',\n",
    "     outfile=f'{test_name}.sdf',\n",
    "     ref_mol_names='x0140_1', # unless alread present as ref_mols column\n",
    "     ref_pdb_name='x0140_1',  # ditto\n",
    "     extras=['monthly_views']  # extras\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "07aa2289-2ccf-4d71-aa07-d1c387886c13",
   "metadata": {},
   "outputs": [],
   "source": [
    "import operator\n",
    "compounds['title'] = compounds.title.str.replace(r'\\W', '_', regex=True).apply(operator.itemgetter(slice(None, 20)))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "778981d5-9146-479a-90a6-86b3bdc3baf0",
   "metadata": {},
   "source": [
    "# Test 2: Words\n",
    "\n",
    "Words"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "0432c94b-b605-45de-8cb9-275b4955f3e8",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[12:45:46] Molecule does not have explicit Hs. Consider calling AddHs()\n"
     ]
    }
   ],
   "source": [
    "test_name = 'MF-test231113_words'\n",
    "\n",
    "header: Chem.Mol = generate_header(method=test_name,\n",
    "                         ref_url='https://www.example.com',\n",
    "                         submitter_name='matteo',\n",
    "                         submitter_email='matteo.ferla@stats.ox.ac.uk',\n",
    "                         submitter_institution='Ox',\n",
    "                         extras={'wordy_views': 'monthly views in Wikipedia', 'monthly_views':'monthly views in Wikipedia'})\n",
    "\n",
    "\n",
    "compounds['wordy_views'] = compounds.monthly_views.apply('Viewed {} times a month'.format)\n",
    "\n",
    "prep(compounds, header, \n",
    "     mol_col='mol', \n",
    "     name_col='title',\n",
    "     outfile=f'{test_name}.sdf',\n",
    "     ref_mol_names='x0140_1', # unless alread present as ref_mols column\n",
    "     ref_pdb_name='x0140_1',  # ditto\n",
    "     extras=['wordy_views', 'monthly_views'],  # extras\n",
    "    )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "f5013201-a0e4-4f01-bb39-3d4a1acfac6d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "254"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(compounds.iloc[34].SMILES)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "bffc3db8-cda5-4e6c-8e18-4b2f6ebc84b3",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:py310]",
   "language": "python",
   "name": "conda-env-py310-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}