{
  "id": 455676,
  "title": "Protein Folding vs RNA Folding",
  "url": "/competitions/stanford-ribonanza-rna-folding/discussion/455676",
  "author_name": "",
  "post_date": "2023-11-15T23:24:27.997459Z",
  "votes": 1,
  "comment_count": 2,
  "views": 0,
  "content": "<p>Hi Kaggle,</p>\n<p>Naive question here from a non-biology data scientist. Looking at a lot of the other related competitions and open source models people have shared (e.g., AlphaFoldv2 on the CASP competition), many of these models and competitions relate to the problem of protein folding, rather than RNA folding (which we have here). While I understand the difference between the two on a biological level, is there a benefit in terms of medicine to being able to predict RNA folding structure if protein folding has already been \"solved\" by AlphaFoldv2? </p>\n<p>Thanks!</p>",
  "messages": [
    {
      "id": "2526594",
      "postDate": "11/15/2023 23:24:27",
      "content": "<p>Hi Kaggle,</p>\n<p>Naive question here from a non-biology data scientist. Looking at a lot of the other related competitions and open source models people have shared (e.g., AlphaFoldv2 on the CASP competition), many of these models and competitions relate to the problem of protein folding, rather than RNA folding (which we have here). While I understand the difference between the two on a biological level, is there a benefit in terms of medicine to being able to predict RNA folding structure if protein folding has already been \"solved\" by AlphaFoldv2? </p>\n<p>Thanks!</p>",
      "rawMarkdown": "Hi Kaggle,\n\nNaive question here from a non-biology data scientist. Looking at a lot of the other related competitions and open source models people have shared (e.g., AlphaFoldv2 on the CASP competition), many of these models and competitions relate to the problem of protein folding, rather than RNA folding (which we have here). While I understand the difference between the two on a biological level, is there a benefit in terms of medicine to being able to predict RNA folding structure if protein folding has already been \"solved\" by AlphaFoldv2? \n\nThanks!",
      "votes": null
    },
    {
      "id": "2526793",
      "postDate": "11/16/2023 03:55:24",
      "content": "<p>It seems alphafold is only confident about 100,000. It seems that since protein is a derivative of amino acids and amino acids are derivatives of DNA (RNA is also employed) we would want to pursue RNA analysis first. Both are required and AlphaFoldv2 analysis will shed light into the progress of both. If I were to hazard a guess I would think it is because RNA has a much broader range of things that can affect its structure. It has a primary and a secondary structure. That aside I wonder if it's because of the paucity of experimental data. If we were to check I wonder if the DMS and 2A3 experiments come after the AlphaFold project. That work maybe shepherds this work.</p>",
      "rawMarkdown": "It seems alphafold is only confident about 100,000. It seems that since protein is a derivative of amino acids and amino acids are derivatives of DNA (RNA is also employed) we would want to pursue RNA analysis first. Both are required and AlphaFoldv2 analysis will shed light into the progress of both. If I were to hazard a guess I would think it is because RNA has a much broader range of things that can affect its structure. It has a primary and a secondary structure. That aside I wonder if it's because of the paucity of experimental data. If we were to check I wonder if the DMS and 2A3 experiments come after the AlphaFold project. That work maybe shepherds this work.",
      "votes": null
    },
    {
      "id": "2529208",
      "postDate": "11/18/2023 03:43:52",
      "content": "<p>Oh, that makes sense. Thanks!</p>",
      "rawMarkdown": "Oh, that makes sense. Thanks!",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 2526793,
      "author_name": "tuttlen",
      "author_url": "",
      "post_date": "11/16/2023 03:55:24",
      "content": "<p>It seems alphafold is only confident about 100,000. It seems that since protein is a derivative of amino acids and amino acids are derivatives of DNA (RNA is also employed) we would want to pursue RNA analysis first. Both are required and AlphaFoldv2 analysis will shed light into the progress of both. If I were to hazard a guess I would think it is because RNA has a much broader range of things that can affect its structure. It has a primary and a secondary structure. That aside I wonder if it's because of the paucity of experimental data. If we were to check I wonder if the DMS and 2A3 experiments come after the AlphaFold project. That work maybe shepherds this work.</p>",
      "votes": null,
      "replies": [
        {
          "id": 2529208,
          "author_name": "alberteinsten",
          "author_url": "",
          "post_date": "11/18/2023 03:43:52",
          "content": "<p>Oh, that makes sense. Thanks!</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "2526594": "Hi Kaggle,\n\nNaive question here from a non-biology data scientist. Looking at a lot of the other related competitions and open source models people have shared (e.g., AlphaFoldv2 on the CASP competition), many of these models and competitions relate to the problem of protein folding, rather than RNA folding (which we have here). While I understand the difference between the two on a biological level, is there a benefit in terms of medicine to being able to predict RNA folding structure if protein folding has already been \"solved\" by AlphaFoldv2? \n\nThanks!",
    "2526793": "It seems alphafold is only confident about 100,000. It seems that since protein is a derivative of amino acids and amino acids are derivatives of DNA (RNA is also employed) we would want to pursue RNA analysis first. Both are required and AlphaFoldv2 analysis will shed light into the progress of both. If I were to hazard a guess I would think it is because RNA has a much broader range of things that can affect its structure. It has a primary and a secondary structure. That aside I wonder if it's because of the paucity of experimental data. If we were to check I wonder if the DMS and 2A3 experiments come after the AlphaFold project. That work maybe shepherds this work.",
    "2529208": "Oh, that makes sense. Thanks!"
  },
  "source": "meta"
}