{
  "id": 443229,
  "title": "I want to make sure I define the problem.",
  "url": "/competitions/open-problems-single-cell-perturbations/discussion/443229",
  "author_name": "",
  "post_date": "2023-09-26T06:07:11.070414600Z",
  "votes": 3,
  "comment_count": 1,
  "views": 0,
  "content": "<p>I understand that gene expression for various drugs and donors is given in adata. Is it correct to calculate the gene expression for the drug and donor missing from adata and then calculate the DE (P-value) using the R package limma based on that value? </p>\n<p>Are you going to post how to find DE using limma?</p>",
  "messages": [
    {
      "id": "2456338",
      "postDate": "09/26/2023 06:07:11",
      "content": "<p>I understand that gene expression for various drugs and donors is given in adata. Is it correct to calculate the gene expression for the drug and donor missing from adata and then calculate the DE (P-value) using the R package limma based on that value? </p>\n<p>Are you going to post how to find DE using limma?</p>",
      "rawMarkdown": "I understand that gene expression for various drugs and donors is given in adata. Is it correct to calculate the gene expression for the drug and donor missing from adata and then calculate the DE (P-value) using the R package limma based on that value? \n\nAre you going to post how to find DE using limma?",
      "votes": null
    },
    {
      "id": "2458821",
      "postDate": "09/27/2023 19:59:22",
      "content": "<p>Hi <a href=\"https://www.kaggle.com/kukseungho\" target=\"_blank\">@kukseungho</a> ! You could indeed do this. You could also train a model directly on calculated DE values. We hard at work getting the DE script out, I know it's taking longer than expected. We'll post an update when it's online.</p>",
      "rawMarkdown": "Hi @kukseungho ! You could indeed do this. You could also train a model directly on calculated DE values. We hard at work getting the DE script out, I know it's taking longer than expected. We'll post an update when it's online.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2458821,
      "author_name": "danielburkhardt",
      "author_url": "",
      "post_date": "09/27/2023 19:59:22",
      "content": "<p>Hi <a href=\"https://www.kaggle.com/kukseungho\" target=\"_blank\">@kukseungho</a> ! You could indeed do this. You could also train a model directly on calculated DE values. We hard at work getting the DE script out, I know it's taking longer than expected. We'll post an update when it's online.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2456338": "I understand that gene expression for various drugs and donors is given in adata. Is it correct to calculate the gene expression for the drug and donor missing from adata and then calculate the DE (P-value) using the R package limma based on that value? \n\nAre you going to post how to find DE using limma?",
    "2458821": "Hi @kukseungho ! You could indeed do this. You could also train a model directly on calculated DE values. We hard at work getting the DE script out, I know it's taking longer than expected. We'll post an update when it's online."
  },
  "source": "meta"
}