{
  "id": 491172,
  "title": "Pre-print of 4th place solution",
  "url": "/competitions/cafa-5-protein-function-prediction/discussion/491172",
  "author_name": "ZMC",
  "post_date": "2024-04-04T22:12:17.867000",
  "votes": 3,
  "comment_count": 1,
  "views": 0,
  "content": "<p>Hi everyone, we have released a short article on biorxiv describing our 4th place solution (PROTGOAT) that can be found here : <a href=\"https://www.biorxiv.org/content/10.1101/2024.04.01.587572v1\" target=\"_blank\">https://www.biorxiv.org/content/10.1101/2024.04.01.587572v1</a> </p>\n<p>Feel free to ask us any questions you might have about our model! </p>",
  "messages": [
    {
      "id": 2735761,
      "postDate": "2024-04-04T22:12:17.867Z",
      "content": "<p>Hi everyone, we have released a short article on biorxiv describing our 4th place solution (PROTGOAT) that can be found here : <a href=\"https://www.biorxiv.org/content/10.1101/2024.04.01.587572v1\" target=\"_blank\">https://www.biorxiv.org/content/10.1101/2024.04.01.587572v1</a> </p>\n<p>Feel free to ask us any questions you might have about our model! </p>",
      "rawMarkdown": "Hi everyone, we have released a short article on biorxiv describing our 4th place solution (PROTGOAT) that can be found here : https://www.biorxiv.org/content/10.1101/2024.04.01.587572v1 \n\nFeel free to ask us any questions you might have about our model! ",
      "votes": 3
    },
    {
      "id": 2774277,
      "postDate": "2024-04-25T05:57:14.530Z",
      "content": "<p>Congratulations for securing fourth place in the leader board. I have a small doubt, in the given test dataset there are many Protein Ids repeating from training dataset, should we need to consider those proteins also during prediction because we already trained that. Another question is that for the new proteins from test data there is GO terms present in the train_terms.tsv file. so when predicting how we can tell that this protein correponds to the predicted GO term.</p>",
      "rawMarkdown": "Congratulations for securing fourth place in the leader board. I have a small doubt, in the given test dataset there are many Protein Ids repeating from training dataset, should we need to consider those proteins also during prediction because we already trained that. Another question is that for the new proteins from test data there is GO terms present in the train_terms.tsv file. so when predicting how we can tell that this protein correponds to the predicted GO term.",
      "isDeleted": true
    }
  ],
  "comments": [
    {
      "id": 2774277,
      "author_name": "",
      "author_url": "",
      "post_date": "2024-04-25T05:57:14.530000",
      "content": "<p>Congratulations for securing fourth place in the leader board. I have a small doubt, in the given test dataset there are many Protein Ids repeating from training dataset, should we need to consider those proteins also during prediction because we already trained that. Another question is that for the new proteins from test data there is GO terms present in the train_terms.tsv file. so when predicting how we can tell that this protein correponds to the predicted GO term.</p>",
      "votes": 0,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2735761": "Hi everyone, we have released a short article on biorxiv describing our 4th place solution (PROTGOAT) that can be found here : https://www.biorxiv.org/content/10.1101/2024.04.01.587572v1 \n\nFeel free to ask us any questions you might have about our model! ",
    "2774277": "Congratulations for securing fourth place in the leader board. I have a small doubt, in the given test dataset there are many Protein Ids repeating from training dataset, should we need to consider those proteins also during prediction because we already trained that. Another question is that for the new proteins from test data there is GO terms present in the train_terms.tsv file. so when predicting how we can tell that this protein correponds to the predicted GO term."
  }
}