{
  "id": 347813,
  "title": "Did time feature helped？",
  "url": "/competitions/open-problems-multimodal/discussion/347813",
  "author_name": "",
  "post_date": "2022-08-25T13:40:27.218254700Z",
  "votes": 2,
  "comment_count": 2,
  "views": 0,
  "content": "<p>As we know that each cell are tested in one day and discharged, which means that there is now time time series relationship between thses cells, should we used time features ?</p>",
  "messages": [
    {
      "id": "1913753",
      "postDate": "08/25/2022 13:40:27",
      "content": "<p>As we know that each cell are tested in one day and discharged, which means that there is now time time series relationship between thses cells, should we used time features ?</p>",
      "rawMarkdown": "As we know that each cell are tested in one day and discharged, which means that there is now time time series relationship between thses cells, should we used time features ?",
      "votes": null
    },
    {
      "id": "1913927",
      "postDate": "08/25/2022 15:26:38",
      "content": "<p>Hi <a href=\"https://www.kaggle.com/leehann\" target=\"_blank\">@leehann</a>! You are right, to measure the cells with CITEseq or Multiome, they have to be lysed (\"destroyed\"). Therefore, you do not have a proper time series for a single cell (a cell is only measured once for a given day). This is part of the challenge. Still, time significantly alters expression of RNA and surface protein expression. So it might be useful to use this information in your model.</p>",
      "rawMarkdown": "Hi @leehann! You are right, to measure the cells with CITEseq or Multiome, they have to be lysed (\"destroyed\"). Therefore, you do not have a proper time series for a single cell (a cell is only measured once for a given day). This is part of the challenge. Still, time significantly alters expression of RNA and surface protein expression. So it might be useful to use this information in your model.",
      "votes": null
    },
    {
      "id": "1914294",
      "postDate": "08/26/2022 00:44:55",
      "content": "<p>thanks so much for your reply</p>",
      "rawMarkdown": "thanks so much for your reply",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1913927,
      "author_name": "peterholderrieth",
      "author_url": "",
      "post_date": "08/25/2022 15:26:38",
      "content": "<p>Hi <a href=\"https://www.kaggle.com/leehann\" target=\"_blank\">@leehann</a>! You are right, to measure the cells with CITEseq or Multiome, they have to be lysed (\"destroyed\"). Therefore, you do not have a proper time series for a single cell (a cell is only measured once for a given day). This is part of the challenge. Still, time significantly alters expression of RNA and surface protein expression. So it might be useful to use this information in your model.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1914294,
      "author_name": "leehann",
      "author_url": "",
      "post_date": "08/26/2022 00:44:55",
      "content": "<p>thanks so much for your reply</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1913753": "As we know that each cell are tested in one day and discharged, which means that there is now time time series relationship between thses cells, should we used time features ?",
    "1913927": "Hi @leehann! You are right, to measure the cells with CITEseq or Multiome, they have to be lysed (\"destroyed\"). Therefore, you do not have a proper time series for a single cell (a cell is only measured once for a given day). This is part of the challenge. Still, time significantly alters expression of RNA and surface protein expression. So it might be useful to use this information in your model.",
    "1914294": "thanks so much for your reply"
  },
  "source": "meta"
}