{
  "id": 640452,
  "title": "Official clarification needed: Is demographic data (age, sex, race, ethnicity) available and usable in the TEST data for ANY group?",
  "url": "/competitions/adaptive-immune-profiling-challenge-2025/discussion/640452",
  "author_name": "",
  "post_date": "2025-11-25T18:37:25.302178Z",
  "votes": 1,
  "comment_count": 5,
  "views": 0,
  "content": "<p>Dear Hosts,</p>\n<p>I respectfully request a clear and official answer regarding <strong>demographic variables only</strong> (age, sex, race, ethnicity):</p>\n<ul>\n<li><p>In the <strong>training</strong> sets, several groups (especially Group 7, and 8) contain strong demographic signals in metadata.csv  </p></li>\n<li><p>However, in the <strong>test</strong> folders currently visible:</p>\n<ul>\n<li>Test_dataset  have <strong>NO age/sex/race/ethnicity columns</strong></li></ul></li>\n</ul>",
  "messages": [
    {
      "id": "3348330",
      "postDate": "11/25/2025 18:37:25",
      "content": "<p>Dear Hosts,</p>\n<p>I respectfully request a clear and official answer regarding <strong>demographic variables only</strong> (age, sex, race, ethnicity):</p>\n<ul>\n<li><p>In the <strong>training</strong> sets, several groups (especially Group 7, and 8) contain strong demographic signals in metadata.csv  </p></li>\n<li><p>However, in the <strong>test</strong> folders currently visible:</p>\n<ul>\n<li>Test_dataset  have <strong>NO age/sex/race/ethnicity columns</strong></li></ul></li>\n</ul>",
      "rawMarkdown": "Dear Hosts,\n\nI respectfully request a clear and official answer regarding **demographic variables only** (age, sex, race, ethnicity):\n\n- In the **training** sets, several groups (especially Group 7, and 8) contain strong demographic signals in metadata.csv  \n\n- However, in the **test** folders currently visible:\n  - Test_dataset  have **NO age/sex/race/ethnicity columns**",
      "votes": null
    },
    {
      "id": "3350079",
      "postDate": "11/27/2025 09:57:56",
      "content": "<p>Good observation! It was a deliberate choice that we did not provide demographic information for test data, where available. The reason is the following: we envision that participants learn generalizable signals reflected in immune repertoire data (the sequence data provided) while accounting for possible biases that could creep in from demographic variables, if any. </p>",
      "rawMarkdown": "Good observation! It was a deliberate choice that we did not provide demographic information for test data, where available. The reason is the following: we envision that participants learn generalizable signals reflected in immune repertoire data (the sequence data provided) while accounting for possible biases that could creep in from demographic variables, if any.",
      "votes": null
    },
    {
      "id": "3350433",
      "postDate": "11/27/2025 14:58:05",
      "content": "<p>What mean this really the public data is only 1% \"This leaderboard is calculated with approximately 1% of the test data. The final results will be based on the other 99%, so the final standings may be different.\"</p>",
      "rawMarkdown": "What mean this really the public data is only 1% \"This leaderboard is calculated with approximately 1% of the test data. The final results will be based on the other 99%, so the final standings may be different.\"",
      "votes": null
    },
    {
      "id": "3350743",
      "postDate": "11/27/2025 18:47:04",
      "content": "<p>Hi! I am not really sure if I follow your question 100%. Could you please elaborate on what you are wondering about?</p>",
      "rawMarkdown": "Hi! I am not really sure if I follow your question 100%. Could you please elaborate on what you are wondering about?",
      "votes": null
    },
    {
      "id": "3350790",
      "postDate": "11/27/2025 19:42:05",
      "content": "<p>On the leaderboard page it saying This leaderboard is calculated with approximately 1% of the test data. The final results will be based on the other 99%, so the final standings may be different.\" Is that means current public score is only calculated on 1% test data ? Or wrong</p>",
      "rawMarkdown": "On the leaderboard page it saying This leaderboard is calculated with approximately 1% of the test data. The final results will be based on the other 99%, so the final standings may be different.\" Is that means current public score is only calculated on 1% test data ? Or wrong",
      "votes": null
    },
    {
      "id": "3351318",
      "postDate": "11/28/2025 10:21:02",
      "content": "<p>That statement is not wrong per se 😀. The percentages you see reflect the fact that continuous feedback is not provided on task 2 (referring to \"tasks\" described on the overview page images) in the Public leaderboard, as mentioned in other threads on the discussion forum. </p>",
      "rawMarkdown": "That statement is not wrong per se 😀. The percentages you see reflect the fact that continuous feedback is not provided on task 2 (referring to \"tasks\" described on the overview page images) in the Public leaderboard, as mentioned in other threads on the discussion forum.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 3350079,
      "author_name": "ckanduri",
      "author_url": "",
      "post_date": "11/27/2025 09:57:56",
      "content": "<p>Good observation! It was a deliberate choice that we did not provide demographic information for test data, where available. The reason is the following: we envision that participants learn generalizable signals reflected in immune repertoire data (the sequence data provided) while accounting for possible biases that could creep in from demographic variables, if any. </p>",
      "votes": null,
      "replies": [
        {
          "id": 3350433,
          "author_name": "yisberh",
          "author_url": "",
          "post_date": "11/27/2025 14:58:05",
          "content": "<p>What mean this really the public data is only 1% \"This leaderboard is calculated with approximately 1% of the test data. The final results will be based on the other 99%, so the final standings may be different.\"</p>",
          "votes": null,
          "replies": [
            {
              "id": 3350743,
              "author_name": "ckanduri",
              "author_url": "",
              "post_date": "11/27/2025 18:47:04",
              "content": "<p>Hi! I am not really sure if I follow your question 100%. Could you please elaborate on what you are wondering about?</p>",
              "votes": null,
              "replies": [
                {
                  "id": 3350790,
                  "author_name": "yisberh",
                  "author_url": "",
                  "post_date": "11/27/2025 19:42:05",
                  "content": "<p>On the leaderboard page it saying This leaderboard is calculated with approximately 1% of the test data. The final results will be based on the other 99%, so the final standings may be different.\" Is that means current public score is only calculated on 1% test data ? Or wrong</p>",
                  "votes": null,
                  "replies": [
                    {
                      "id": 3351318,
                      "author_name": "ckanduri",
                      "author_url": "",
                      "post_date": "11/28/2025 10:21:02",
                      "content": "<p>That statement is not wrong per se 😀. The percentages you see reflect the fact that continuous feedback is not provided on task 2 (referring to \"tasks\" described on the overview page images) in the Public leaderboard, as mentioned in other threads on the discussion forum. </p>",
                      "votes": null,
                      "replies": []
                    }
                  ]
                }
              ]
            }
          ]
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "3348330": "Dear Hosts,\n\nI respectfully request a clear and official answer regarding **demographic variables only** (age, sex, race, ethnicity):\n\n- In the **training** sets, several groups (especially Group 7, and 8) contain strong demographic signals in metadata.csv  \n\n- However, in the **test** folders currently visible:\n  - Test_dataset  have **NO age/sex/race/ethnicity columns**",
    "3350079": "Good observation! It was a deliberate choice that we did not provide demographic information for test data, where available. The reason is the following: we envision that participants learn generalizable signals reflected in immune repertoire data (the sequence data provided) while accounting for possible biases that could creep in from demographic variables, if any.",
    "3350433": "What mean this really the public data is only 1% \"This leaderboard is calculated with approximately 1% of the test data. The final results will be based on the other 99%, so the final standings may be different.\"",
    "3350743": "Hi! I am not really sure if I follow your question 100%. Could you please elaborate on what you are wondering about?",
    "3350790": "On the leaderboard page it saying This leaderboard is calculated with approximately 1% of the test data. The final results will be based on the other 99%, so the final standings may be different.\" Is that means current public score is only calculated on 1% test data ? Or wrong",
    "3351318": "That statement is not wrong per se 😀. The percentages you see reflect the fact that continuous feedback is not provided on task 2 (referring to \"tasks\" described on the overview page images) in the Public leaderboard, as mentioned in other threads on the discussion forum."
  },
  "source": "meta"
}