{
  "id": 443686,
  "title": "question about target features",
  "url": "/competitions/open-problems-single-cell-perturbations/discussion/443686",
  "author_name": "",
  "post_date": "2023-09-28T09:19:28.319471Z",
  "votes": 4,
  "comment_count": 3,
  "views": 0,
  "content": "<p>In the data tab, we can see the following descriptions: </p>\n<p>(1) \"Note that there is no additional test data beyond the indicated cell_type / sm_name pairs. The input to your model will be a tuple of cell_type and sm_name and the output of your model will be predicted <strong>signed -log10(p-values)</strong> for all 18211 genes.\"</p>\n<p>It means we will predict: <strong>signed * -log10(p-values)</strong>, right? </p>\n<p>(2) \"genes A1BG, A1BG-AS1, …, ZZEF1 (numbering 18,211 in all) - Differential expression value <strong>(-log10(p-value) * sign(LFC))</strong> for each gene. \"</p>\n<p>What is LFC? I cannot find anything related to this term in the description, please clarify it.</p>",
  "messages": [
    {
      "id": "2459557",
      "postDate": "09/28/2023 09:19:28",
      "content": "<p>In the data tab, we can see the following descriptions: </p>\n<p>(1) \"Note that there is no additional test data beyond the indicated cell_type / sm_name pairs. The input to your model will be a tuple of cell_type and sm_name and the output of your model will be predicted <strong>signed -log10(p-values)</strong> for all 18211 genes.\"</p>\n<p>It means we will predict: <strong>signed * -log10(p-values)</strong>, right? </p>\n<p>(2) \"genes A1BG, A1BG-AS1, …, ZZEF1 (numbering 18,211 in all) - Differential expression value <strong>(-log10(p-value) * sign(LFC))</strong> for each gene. \"</p>\n<p>What is LFC? I cannot find anything related to this term in the description, please clarify it.</p>",
      "rawMarkdown": "In the data tab, we can see the following descriptions: \n\n(1) \"Note that there is no additional test data beyond the indicated cell_type / sm_name pairs. The input to your model will be a tuple of cell_type and sm_name and the output of your model will be predicted **signed -log10(p-values)** for all 18211 genes.\"\n\nIt means we will predict: **signed * -log10(p-values)**, right? \n\n\n(2) \"genes A1BG, A1BG-AS1, …, ZZEF1 (numbering 18,211 in all) - Differential expression value **(-log10(p-value) * sign(LFC))** for each gene. \"\n\nWhat is LFC? I cannot find anything related to this term in the description, please clarify it.",
      "votes": null
    },
    {
      "id": "2460092",
      "postDate": "09/28/2023 15:11:35",
      "content": "<p>LFC stands for Log Fold Change and it is usually reported by differential expression analysis tool, they report the effect size and direction of change of genes (typically treated vs control). </p>\n<p>In this challenge, organizers decided to report only the statistical significance of the change, but not the effect size (the LFC). So the value you need to predict is basically telling you how significant is the change in gene expression value for a given compound, and in which direction changed, but you don't know if the change was large or small</p>",
      "rawMarkdown": "LFC stands for Log Fold Change and it is usually reported by differential expression analysis tool, they report the effect size and direction of change of genes (typically treated vs control). \n\nIn this challenge, organizers decided to report only the statistical significance of the change, but not the effect size (the LFC). So the value you need to predict is basically telling you how significant is the change in gene expression value for a given compound, and in which direction changed, but you don't know if the change was large or small",
      "votes": null
    },
    {
      "id": "2460104",
      "postDate": "09/28/2023 15:20:32",
      "content": "<p>Thanks for the feedback! I added text in the <code>data</code> description to clarify:</p>\n<blockquote>\n  <p>genes A1BG, A1BG-AS1, …, ZZEF1 (numbering 18,211 in all) - Differential expression value (-log10(p-value) * sign(LFC)) for each gene. Here, LFC is the estimated log-fold change in expression between the treatment and control condition after shrinkage as calculated by Limma. Positive LFC means the gene goes up in the treatment condition relative to the control.</p>\n</blockquote>\n<p>Does this make sense? Happy to provide more context</p>",
      "rawMarkdown": "Thanks for the feedback! I added text in the `data` description to clarify:\n\n> genes A1BG, A1BG-AS1, …, ZZEF1 (numbering 18,211 in all) - Differential expression value (-log10(p-value) * sign(LFC)) for each gene. Here, LFC is the estimated log-fold change in expression between the treatment and control condition after shrinkage as calculated by Limma. Positive LFC means the gene goes up in the treatment condition relative to the control.\n\nDoes this make sense? Happy to provide more context",
      "votes": null
    },
    {
      "id": "2461090",
      "postDate": "09/29/2023 09:29:07",
      "content": "<p>Thanks <a href=\"https://www.kaggle.com/danielburkhardt\" target=\"_blank\">@danielburkhardt</a> , <a href=\"https://www.kaggle.com/pablormier\" target=\"_blank\">@pablormier</a>  for your feedback!</p>",
      "rawMarkdown": "Thanks @danielburkhardt , @pablormier  for your feedback!",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2460092,
      "author_name": "pablormier",
      "author_url": "",
      "post_date": "09/28/2023 15:11:35",
      "content": "<p>LFC stands for Log Fold Change and it is usually reported by differential expression analysis tool, they report the effect size and direction of change of genes (typically treated vs control). </p>\n<p>In this challenge, organizers decided to report only the statistical significance of the change, but not the effect size (the LFC). So the value you need to predict is basically telling you how significant is the change in gene expression value for a given compound, and in which direction changed, but you don't know if the change was large or small</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 2460104,
      "author_name": "danielburkhardt",
      "author_url": "",
      "post_date": "09/28/2023 15:20:32",
      "content": "<p>Thanks for the feedback! I added text in the <code>data</code> description to clarify:</p>\n<blockquote>\n  <p>genes A1BG, A1BG-AS1, …, ZZEF1 (numbering 18,211 in all) - Differential expression value (-log10(p-value) * sign(LFC)) for each gene. Here, LFC is the estimated log-fold change in expression between the treatment and control condition after shrinkage as calculated by Limma. Positive LFC means the gene goes up in the treatment condition relative to the control.</p>\n</blockquote>\n<p>Does this make sense? Happy to provide more context</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 2461090,
      "author_name": "mathormad",
      "author_url": "",
      "post_date": "09/29/2023 09:29:07",
      "content": "<p>Thanks <a href=\"https://www.kaggle.com/danielburkhardt\" target=\"_blank\">@danielburkhardt</a> , <a href=\"https://www.kaggle.com/pablormier\" target=\"_blank\">@pablormier</a>  for your feedback!</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2459557": "In the data tab, we can see the following descriptions: \n\n(1) \"Note that there is no additional test data beyond the indicated cell_type / sm_name pairs. The input to your model will be a tuple of cell_type and sm_name and the output of your model will be predicted **signed -log10(p-values)** for all 18211 genes.\"\n\nIt means we will predict: **signed * -log10(p-values)**, right? \n\n\n(2) \"genes A1BG, A1BG-AS1, …, ZZEF1 (numbering 18,211 in all) - Differential expression value **(-log10(p-value) * sign(LFC))** for each gene. \"\n\nWhat is LFC? I cannot find anything related to this term in the description, please clarify it.",
    "2460092": "LFC stands for Log Fold Change and it is usually reported by differential expression analysis tool, they report the effect size and direction of change of genes (typically treated vs control). \n\nIn this challenge, organizers decided to report only the statistical significance of the change, but not the effect size (the LFC). So the value you need to predict is basically telling you how significant is the change in gene expression value for a given compound, and in which direction changed, but you don't know if the change was large or small",
    "2460104": "Thanks for the feedback! I added text in the `data` description to clarify:\n\n> genes A1BG, A1BG-AS1, …, ZZEF1 (numbering 18,211 in all) - Differential expression value (-log10(p-value) * sign(LFC)) for each gene. Here, LFC is the estimated log-fold change in expression between the treatment and control condition after shrinkage as calculated by Limma. Positive LFC means the gene goes up in the treatment condition relative to the control.\n\nDoes this make sense? Happy to provide more context",
    "2461090": "Thanks @danielburkhardt , @pablormier  for your feedback!"
  },
  "source": "meta"
}