{
  "id": 453373,
  "title": "SMILES for testset (i.e. id_map)",
  "url": "/competitions/open-problems-single-cell-perturbations/discussion/453373",
  "author_name": "",
  "post_date": "2023-11-06T05:47:49.258178200Z",
  "votes": 1,
  "comment_count": 2,
  "views": 0,
  "content": "<p>There is no SMILES for the id_map. But I am using SMILES for prediction. I tried using rdkit to convert the sm_name to SMILES, but I failed.</p>\n<p>Can any one convert the name to SMILES?</p>",
  "messages": [
    {
      "id": "2514196",
      "postDate": "11/06/2023 05:47:49",
      "content": "<p>There is no SMILES for the id_map. But I am using SMILES for prediction. I tried using rdkit to convert the sm_name to SMILES, but I failed.</p>\n<p>Can any one convert the name to SMILES?</p>",
      "rawMarkdown": "There is no SMILES for the id_map. But I am using SMILES for prediction. I tried using rdkit to convert the sm_name to SMILES, but I failed.\n\nCan any one convert the name to SMILES?",
      "votes": null
    },
    {
      "id": "2514231",
      "postDate": "11/06/2023 06:22:09",
      "content": "<p>Hi! You can find all test drugs SMILES in the de_train since all the test compounds are there. </p>",
      "rawMarkdown": "Hi! You can find all test drugs SMILES in the de_train since all the test compounds are there.",
      "votes": null
    },
    {
      "id": "2514699",
      "postDate": "11/06/2023 13:16:56",
      "content": "<p>You can build a dataframe of the sm_name and SMILES from the train data - should end up with 146 rows.  Than merge that dataframe into test, as there are no NEW compounds in test.</p>",
      "rawMarkdown": "You can build a dataframe of the sm_name and SMILES from the train data - should end up with 146 rows.  Than merge that dataframe into test, as there are no NEW compounds in test.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2514231,
      "author_name": "antoninadolgorukova",
      "author_url": "",
      "post_date": "11/06/2023 06:22:09",
      "content": "<p>Hi! You can find all test drugs SMILES in the de_train since all the test compounds are there. </p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 2514699,
      "author_name": "pcjimmmy",
      "author_url": "",
      "post_date": "11/06/2023 13:16:56",
      "content": "<p>You can build a dataframe of the sm_name and SMILES from the train data - should end up with 146 rows.  Than merge that dataframe into test, as there are no NEW compounds in test.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2514196": "There is no SMILES for the id_map. But I am using SMILES for prediction. I tried using rdkit to convert the sm_name to SMILES, but I failed.\n\nCan any one convert the name to SMILES?",
    "2514231": "Hi! You can find all test drugs SMILES in the de_train since all the test compounds are there.",
    "2514699": "You can build a dataframe of the sm_name and SMILES from the train data - should end up with 146 rows.  Than merge that dataframe into test, as there are no NEW compounds in test."
  },
  "source": "meta"
}