{
  "id": 671575,
  "title": " Chains and ligands",
  "url": "/competitions/stanford-rna-3d-folding-2/discussion/671575",
  "author_name": "Sofiya Rachkevych",
  "post_date": "2026-02-02T19:42:12.439000",
  "votes": 4,
  "comment_count": 1,
  "views": 0,
  "content": "<p>Okay, what is this chain thing? If it's a RNA chain, why then I see both A and B in the same RNA molecule in  the train_labels.csv? \nAlso ligand information: is it something nice-to-have?\nIt wasn't last year so just asking :)</p>",
  "messages": [
    {
      "id": 3401584,
      "postDate": "2026-02-04T04:17:00.963Z",
      "content": "<p><strong>Chains</strong></p>\n<p>In this Kaggle competition,  biological assemblies are not separated into individual chains. Instead, the RNA composition of the assembly is recorded in <code>stoichiometry</code>.</p>\n<p>For example, the target ID <a href=\"https://www.rcsb.org/structure/353D\" target=\"_blank\">353D</a> has two distinct RNA chains, A and B, in its biological assembly. This target's overall <code>sequence</code> is the concatenation of chains A and B, and its <code>stoichiometry</code> is given as <code>A:1;B:1</code>. The individual sequences for chains A and B are included in <code>all_sequences</code> (we use the <em>auth</em> chain IDs for stoichiometry).</p>\n<p>Correction regarding <code>target_id</code>: Please note that there is no <code>chain_id</code> encoded in the <code>target_id</code> anymore. We apologize for this confusion.</p>\n<p><strong>Ligands</strong></p>\n<p>Ligands and, if present, proteins (as documented in <code>all_sequences</code>), can affect RNA conformation and potentially simplify prediction. Therefore, we recommend to include them during the prediction process, provided the method supports it. However, only the RNA component should be included in your final submission.</p>",
      "rawMarkdown": "**Chains**\n\nIn this Kaggle competition,  biological assemblies are not separated into individual chains. Instead, the RNA composition of the assembly is recorded in `stoichiometry`.\n\nFor example, the target ID [353D](https://www.rcsb.org/structure/353D) has two distinct RNA chains, A and B, in its biological assembly. This target's overall `sequence` is the concatenation of chains A and B, and its `stoichiometry` is given as `A:1;B:1`. The individual sequences for chains A and B are included in `all_sequences` (we use the _auth_ chain IDs for stoichiometry).\n\nCorrection regarding `target_id`: Please note that there is no `chain_id` encoded in the `target_id` anymore. We apologize for this confusion.\n\n**Ligands**\n\nLigands and, if present, proteins (as documented in `all_sequences`), can affect RNA conformation and potentially simplify prediction. Therefore, we recommend to include them during the prediction process, provided the method supports it. However, only the RNA component should be included in your final submission.",
      "votes": 3
    },
    {
      "id": 3401034,
      "postDate": "2026-02-02T19:42:12.440Z",
      "content": "<p>Okay, what is this chain thing? If it's a RNA chain, why then I see both A and B in the same RNA molecule in  the train_labels.csv? \nAlso ligand information: is it something nice-to-have?\nIt wasn't last year so just asking :)</p>",
      "rawMarkdown": "Okay, what is this chain thing? If it's a RNA chain, why then I see both A and B in the same RNA molecule in  the train_labels.csv? \nAlso ligand information: is it something nice-to-have?\nIt wasn't last year so just asking :)",
      "votes": 4
    }
  ],
  "comments": [
    {
      "id": 3401584,
      "author_name": "Przemek Porebski",
      "author_url": "",
      "post_date": "2026-02-04T04:17:00.963000",
      "content": "<p><strong>Chains</strong></p>\n<p>In this Kaggle competition,  biological assemblies are not separated into individual chains. Instead, the RNA composition of the assembly is recorded in <code>stoichiometry</code>.</p>\n<p>For example, the target ID <a href=\"https://www.rcsb.org/structure/353D\" target=\"_blank\">353D</a> has two distinct RNA chains, A and B, in its biological assembly. This target's overall <code>sequence</code> is the concatenation of chains A and B, and its <code>stoichiometry</code> is given as <code>A:1;B:1</code>. The individual sequences for chains A and B are included in <code>all_sequences</code> (we use the <em>auth</em> chain IDs for stoichiometry).</p>\n<p>Correction regarding <code>target_id</code>: Please note that there is no <code>chain_id</code> encoded in the <code>target_id</code> anymore. We apologize for this confusion.</p>\n<p><strong>Ligands</strong></p>\n<p>Ligands and, if present, proteins (as documented in <code>all_sequences</code>), can affect RNA conformation and potentially simplify prediction. Therefore, we recommend to include them during the prediction process, provided the method supports it. However, only the RNA component should be included in your final submission.</p>",
      "votes": 3,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "3401584": "**Chains**\n\nIn this Kaggle competition,  biological assemblies are not separated into individual chains. Instead, the RNA composition of the assembly is recorded in `stoichiometry`.\n\nFor example, the target ID [353D](https://www.rcsb.org/structure/353D) has two distinct RNA chains, A and B, in its biological assembly. This target's overall `sequence` is the concatenation of chains A and B, and its `stoichiometry` is given as `A:1;B:1`. The individual sequences for chains A and B are included in `all_sequences` (we use the _auth_ chain IDs for stoichiometry).\n\nCorrection regarding `target_id`: Please note that there is no `chain_id` encoded in the `target_id` anymore. We apologize for this confusion.\n\n**Ligands**\n\nLigands and, if present, proteins (as documented in `all_sequences`), can affect RNA conformation and potentially simplify prediction. Therefore, we recommend to include them during the prediction process, provided the method supports it. However, only the RNA component should be included in your final submission.",
    "3401034": "Okay, what is this chain thing? If it's a RNA chain, why then I see both A and B in the same RNA molecule in  the train_labels.csv? \nAlso ligand information: is it something nice-to-have?\nIt wasn't last year so just asking :)"
  }
}