Benchmarking machine-readable vectors of chemical reactions on computed activation barriers

JSON Export

{
  "updated": "2024-10-16T11:31:04.970588+00:00", 
  "id": "2401", 
  "metadata": {
    "publication_date": "Oct 16, 2024, 13:31:04", 
    "edited_by": 576, 
    "doi": "10.24435/materialscloud:xd-10", 
    "references": [
      {
        "type": "Journal reference", 
        "url": "https://xlink.rsc.org/?DOI=D3DD00175J", 
        "citation": "P. van Gerwen, K. R. Briling, Y. Calvino Alonso, M. Franke, and C. Corminboeuf, Digital Discovery 3, 932\u2013943 (2024)", 
        "doi": "10.1039/D3DD00175J"
      }
    ], 
    "description": "In recent years, there has been a surge of interest in predicting computed activation barriers, to enable the acceleration of the automated exploration of reaction networks. Consequently, various predictive approaches have emerged, ranging from graph-based models to methods based on the three-dimensional structure of reactants and products. In tandem, many representations have been developed to predict experimental targets, which may hold promise for barrier prediction as well. Here, we bring together all of these efforts and benchmark various methods (Morgan fingerprints, the DRFP, the CGR representation-based Chemprop, SLATM<sub>d</sub>, B\u00b2R<sub>l</sub>\u00b2, EquiReact and language model BERT + RXNFP) for the prediction of computed activation barriers on three diverse datasets.\nThis record includes data to support the article \"Benchmarking machine-readable vectors of chemical reactions on computed activation barriers\". This supports the github repository https://github.com/lcmd-epfl/benchmark-barrier-learning which contains the codes and duplicates the data.", 
    "contributors": [
      {
        "affiliations": [
          "Institut des Sciences et Ing\u00e9nierie Chimiques, \u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne (EPFL), CH-1015 Lausanne, Switzerland", 
          "Laboratory for Computational Molecular Design, Institute of Chemical Sciences and Engineering, \u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne (EPFL), CH-1015 Lausanne, Switzerland", 
          "National Center for Competence in Research-Catalysis (NCCR-Catalysis), \u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne, CH-1015 Lausanne, Switzerland"
        ], 
        "givennames": "Puck", 
        "familyname": "van Gerwen"
      }, 
      {
        "email": "ksenia.briling@epfl.ch", 
        "affiliations": [
          "Institut des Sciences et Ing\u00e9nierie Chimiques, \u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne (EPFL), CH-1015 Lausanne, Switzerland", 
          "Laboratory for Computational Molecular Design, Institute of Chemical Sciences and Engineering, \u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne (EPFL), CH-1015 Lausanne, Switzerland"
        ], 
        "givennames": "Ksenia", 
        "familyname": "R. Briling"
      }, 
      {
        "email": "yannick.calvinoalonso@epfl.ch", 
        "affiliations": [
          "Institut des Sciences et Ing\u00e9nierie Chimiques, \u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne (EPFL), CH-1015 Lausanne, Switzerland", 
          "Laboratory for Computational Molecular Design, Institute of Chemical Sciences and Engineering, \u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne (EPFL), CH-1015 Lausanne, Switzerland"
        ], 
        "givennames": "Yannick", 
        "familyname": "Calvino Alonso"
      }, 
      {
        "affiliations": [
          "Laboratory for Computational Molecular Design, Institute of Chemical Sciences and Engineering, \u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne (EPFL), CH-1015 Lausanne, Switzerland"
        ], 
        "givennames": "Malte", 
        "familyname": "Franke"
      }, 
      {
        "email": "clemence.corminboeuf@epfl.ch", 
        "affiliations": [
          "Institut des Sciences et Ing\u00e9nierie Chimiques, \u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne (EPFL), CH-1015 Lausanne, Switzerland", 
          "Laboratory for Computational Molecular Design, Institute of Chemical Sciences and Engineering, \u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne (EPFL), CH-1015 Lausanne, Switzerland", 
          "National Center for Competence in Research-Catalysis (NCCR-Catalysis), \u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne, CH-1015 Lausanne, Switzerland"
        ], 
        "givennames": "Clemence", 
        "familyname": "Corminboeuf"
      }
    ], 
    "is_last": true, 
    "license_addendum": null, 
    "_files": [
      {
        "description": "The tar ball file `datasets.tar.gz` contains three folders corresponding to each dataset used in the article.\nEach of them contains the geometries (xyz-files), SMILES and properties (CSV-file), and the raw binary data (data-splits, results, and fingerprints/representations)\nSee README.txt for more information.", 
        "checksum": "md5:4a979a671d7fdbcdc99bbe4578c36a0f", 
        "size": 989839360, 
        "key": "datasets.tar.gz"
      }, 
      {
        "description": "README", 
        "checksum": "md5:f9d6e150a6b9bc932fcc7ca2bd535e97", 
        "size": 5336, 
        "key": "README.txt"
      }
    ], 
    "version": 1, 
    "id": "2401", 
    "conceptrecid": "2400", 
    "owner": 1210, 
    "license": "Creative Commons Attribution 4.0 International", 
    "mcid": "2024.163", 
    "_oai": {
      "id": "oai:materialscloud.org:2401"
    }, 
    "status": "published", 
    "title": "Benchmarking machine-readable vectors of chemical reactions on computed activation barriers", 
    "keywords": [
      "chemical reactions", 
      "machine learning", 
      "benchmark", 
      "EPFL", 
      "MARVEL/P2", 
      "SNSF", 
      "ERC"
    ]
  }, 
  "revision": 6, 
  "created": "2024-10-14T16:23:25.251939+00:00"
}