Could you provide the code of how to train the DE model and get the p-value in the training data? I am quite confused.
", "messages": [ { "id": "2436149", "postDate": "09/13/2023 11:38:42", "content": "Could you provide the code of how to train the DE model and get the p-value in the training data? I am quite confused.
", "rawMarkdown": "Could you provide the code of how to train the DE model and get the p-value in the training data? I am quite confused.", "votes": null }, { "id": "2439467", "postDate": "09/14/2023 21:28:17", "content": "I am also confused about how DE analyses were conducted:
\nFor DE analysis, we will need control and treatment. Do we treat column 3 of each plate as controls (DMSO)? Therefore, per DE analysis (one gene and one cell type), we have 3 samples for treatment and 24 samples for control? Also, were DE analyses running on single-cell level or pseudo-bulk level? I saw discrepancies in the description.
\n![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F3619490%2Fb63a3305a66178b94091c82469222436%2FScreen%20Shot%202023-09-14%20at%204.26.52%20PM.png?generation=1694726894715209&alt=media\")
\n![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F3619490%2F97623b6b6eacef7e62c4f6e94a029186%2FScreen%20Shot%202023-09-14%20at%204.26.30%20PM.png?generation=1694726882528961&alt=media\")
Yes, we will provide the code for this! @andrewbenz is working on this, and we should have it up in the next couple days. The challenge is that the code we used relied on our internal repositories and uses R and Python, so getting that to run on Kaggle is a little bit of a challenge. Know that he's working hard to get it done :)
", "rawMarkdown": "Yes, we will provide the code for this! @andrewbenz is working on this, and we should have it up in the next couple days. The challenge is that the code we used relied on our internal repositories and uses R and Python, so getting that to run on Kaggle is a little bit of a challenge. Know that he's working hard to get it done :)", "votes": null }, { "id": "2446754", "postDate": "09/19/2023 15:36:23", "content": "is it already available?
", "rawMarkdown": "is it already available?", "votes": null }, { "id": "2456019", "postDate": "09/25/2023 21:55:15", "content": "@danielburkhardt Has the code been posted? I wasn't able to find it.
", "rawMarkdown": "danielburkhardt Has the code been posted? I wasn't able to find it.", "votes": null } ], "comments": [ { "id": 2439467, "author_name": "taotaotan", "author_url": "", "post_date": "09/14/2023 21:28:17", "content": "I am also confused about how DE analyses were conducted:
\nFor DE analysis, we will need control and treatment. Do we treat column 3 of each plate as controls (DMSO)? Therefore, per DE analysis (one gene and one cell type), we have 3 samples for treatment and 24 samples for control? Also, were DE analyses running on single-cell level or pseudo-bulk level? I saw discrepancies in the description.
\n
\n
Yes, we will provide the code for this! @andrewbenz is working on this, and we should have it up in the next couple days. The challenge is that the code we used relied on our internal repositories and uses R and Python, so getting that to run on Kaggle is a little bit of a challenge. Know that he's working hard to get it done :)
", "votes": null, "replies": [ { "id": 2446754, "author_name": "ioneliabuzatu", "author_url": "", "post_date": "09/19/2023 15:36:23", "content": "is it already available?
", "votes": null, "replies": [] }, { "id": 2456019, "author_name": "danbraswell", "author_url": "", "post_date": "09/25/2023 21:55:15", "content": "@danielburkhardt Has the code been posted? I wasn't able to find it.
", "votes": null, "replies": [] } ] } ] } ], "raw_markdown_by_id": { "2436149": "Could you provide the code of how to train the DE model and get the p-value in the training data? I am quite confused.", "2439467": "I am also confused about how DE analyses were conducted: \nFor DE analysis, we will need control and treatment. Do we treat column 3 of each plate as controls (DMSO)? Therefore, per DE analysis (one gene and one cell type), we have 3 samples for treatment and 24 samples for control? Also, were DE analyses running on single-cell level or pseudo-bulk level? I saw discrepancies in the description.\n\n", "2439473": "Yes, we will provide the code for this! @andrewbenz is working on this, and we should have it up in the next couple days. The challenge is that the code we used relied on our internal repositories and uses R and Python, so getting that to run on Kaggle is a little bit of a challenge. Know that he's working hard to get it done :)", "2446754": "is it already available?", "2456019": "danielburkhardt Has the code been posted? I wasn't able to find it." }, "source": "meta" }