Kramer_AtomInvariants.ipynb

Kramer_AtomInvariants.ipynb

{
  "cells": [
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "from rdkit import Chem\nfrom rdkit.Chem import rdMolDescriptors\nimport rdkit\nprint(rdkit.__version__)",
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "text": "2019.09.2\n",
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": "RDKit WARNING: [13:43:33] Enabling RDKit 2019.09.2 jupyter extensions\n",
          "name": "stderr"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "phenyl = Chem.MolFromSmiles(\"[*:1]c1ccccc1\")\ncyclohexyl = Chem.MolFromSmiles(\"[*:1]C1=CCCCC1\")\ntolyl = Chem.MolFromSmiles(\"[*:1]c1ccccc1C\")",
      "execution_count": 2,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "print(rdMolDescriptors.GetMorganFingerprint(phenyl,0,fromAtoms=[0]).GetNonzeroElements())\nprint(rdMolDescriptors.GetMorganFingerprint(cyclohexyl,0,fromAtoms=[0]).GetNonzeroElements())\nprint(rdMolDescriptors.GetMorganFingerprint(tolyl,0,fromAtoms=[0]).GetNonzeroElements())\n",
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "stream",
          "text": "{2342113506: 1}\n{2342113506: 1}\n{2342113506: 1}\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "def get_simple_morgan(m,radius,includeAromaticity=False,**kwargs):\n    if not includeAromaticity:\n        invars = [x.GetAtomicNum() for x in m.GetAtoms()]\n    else:\n        invars = [x.GetAtomicNum()|(1000+x.GetIsAromatic()) for x in m.GetAtoms()]\n    return rdMolDescriptors.GetMorganFingerprint(m,radius,invariants=invars,**kwargs)",
      "execution_count": 9,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "print(get_simple_morgan(phenyl,1,fromAtoms=[0]).GetNonzeroElements())\nprint(get_simple_morgan(cyclohexyl,1,fromAtoms=[0]).GetNonzeroElements())\nprint(get_simple_morgan(tolyl,1,fromAtoms=[0]).GetNonzeroElements())\n",
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "stream",
          "text": "{0: 1, 3205496007: 1}\n{0: 1, 3205496007: 1}\n{0: 1, 3205496007: 1}\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "print(get_simple_morgan(phenyl,1,includeAromaticity=True,fromAtoms=[0]).GetNonzeroElements())\nprint(get_simple_morgan(cyclohexyl,1,includeAromaticity=True,fromAtoms=[0]).GetNonzeroElements())\nprint(get_simple_morgan(tolyl,1,includeAromaticity=True,fromAtoms=[0]).GetNonzeroElements())",
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "stream",
          "text": "{1000: 1, 3205368108: 1}\n{1000: 1, 3205368113: 1}\n{1000: 1, 3205368108: 1}\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "print(get_simple_morgan(phenyl,2,includeAromaticity=True,fromAtoms=[0]).GetNonzeroElements())\nprint(get_simple_morgan(cyclohexyl,2,includeAromaticity=True,fromAtoms=[0]).GetNonzeroElements())\nprint(get_simple_morgan(tolyl,2,includeAromaticity=True,fromAtoms=[0]).GetNonzeroElements())",
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "stream",
          "text": "{1000: 1, 1524513631: 1, 3205368108: 1}\n{1000: 1, 1525144991: 1, 3205368113: 1}\n{1000: 1, 1524513631: 1, 3205368108: 1}\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "import hashlib\ndef get_smiles_morgan(m,radius,**kwargs):\n    smis = [Chem.Atom.GetSmarts(x) for x in m.GetAtoms()]\n    invars = []\n    for x in m.GetAtoms():\n        # there's almost certainly a more performant way to do this, but....\n        h = hashlib.md5()\n        h.update(x.GetSmarts().encode())\n        invars.append(int.from_bytes(h.digest()[:4],'little'))\n    return rdMolDescriptors.GetMorganFingerprint(m,radius,invariants=invars,**kwargs)",
      "execution_count": 32,
      "outputs": []
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "print(get_smiles_morgan(phenyl,2,fromAtoms=[0]).GetNonzeroElements())\nprint(get_smiles_morgan(cyclohexyl,2,fromAtoms=[0]).GetNonzeroElements())\nprint(get_smiles_morgan(tolyl,2,fromAtoms=[0]).GetNonzeroElements())",
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "stream",
          "text": "{1916351442: 1, 2355797165: 1, 3887229975: 1}\n{1594197318: 1, 1916351442: 1, 3294283874: 1}\n{1916351442: 1, 2355797165: 1, 3887229975: 1}\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {},
      "cell_type": "markdown",
      "source": "Note that this is sensitive to things like atom map numbers:"
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "phenyl2 = Chem.MolFromSmiles(\"[*:2]c1ccccc1\")\nprint(get_smiles_morgan(phenyl2,2,fromAtoms=[0]).GetNonzeroElements())",
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "stream",
          "text": "{1272349123: 1, 1695625860: 1, 2878854048: 1}\n",
          "name": "stdout"
        }
      ]
    },
    {
      "metadata": {
        "trusted": true
      },
      "cell_type": "code",
      "source": "",
      "execution_count": null,
      "outputs": []
    }
  ],
  "metadata": {
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3",
      "language": "python"
    },
    "language_info": {
      "name": "python",
      "version": "3.7.5",
      "mimetype": "text/x-python",
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "pygments_lexer": "ipython3",
      "nbconvert_exporter": "python",
      "file_extension": ".py"
    },
    "gist": {
      "id": "d31ae7618cc5b7322a7121a529bf8190",
      "data": {
        "description": "Kramer_AtomInvariants.ipynb",
        "public": true
      }
    },
    "_draft": {
      "nbviewer_url": "https://gist.github.com/d31ae7618cc5b7322a7121a529bf8190"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 2
}