The data refresh is now complete, and it is great to see that most of the previously entered notebooks were able to generate RNA 3D structure predictions after this stress test. Special thanks to @inversion for coordinating the data refresh!
\nEven more exciting, many of the previously leading notebooks remain highly competitive after the refresh, close to or even exceeding the vfold_human_expert baseline!
The Early Sharing Prizes are now open.
\nIf you are one of the first two to beat thevfold_human_expert baseline and publish a well-documented notebook that respects temporal_cutoff, you get the Early Sharing Prize - and a very special recognition in the history of biology as the author of the first superhuman automated models for 3D RNA structure prediction.
To help us track submissions, it would be helpful to label your Notebook with “Early Sharing Prize” at the beginning of the Notebook title, and to make a post describing your solution herein Discussion. The full Eligibility conditions for the Early Sharing Prize are on our competition overview page.
\nGood luck!
\nFor the hosts,
\nRhiju
The data refresh is now complete, and it is great to see that most of the previously entered notebooks were able to generate RNA 3D structure predictions after this stress test. Special thanks to @inversion for coordinating the data refresh!
\nEven more exciting, many of the previously leading notebooks remain highly competitive after the refresh, close to or even exceeding the vfold_human_expert baseline!
The Early Sharing Prizes are now open.
\nIf you are one of the first two to beat thevfold_human_expert baseline and publish a well-documented notebook that respects temporal_cutoff, you get the Early Sharing Prize - and a very special recognition in the history of biology as the author of the first superhuman automated models for 3D RNA structure prediction.
To help us track submissions, it would be helpful to label your Notebook with “Early Sharing Prize” at the beginning of the Notebook title, and to make a post describing your solution herein Discussion. The full Eligibility conditions for the Early Sharing Prize are on our competition overview page.
\nGood luck!
\nFor the hosts,
\nRhiju
Hello Rhiju et al,
\nAre you able to share the current status of the early sharing prize? I'm wondering if I can find the associated notebooks, but don't see them clearly flagged. I suspect I may be reading the public leaderboard wrong, but it looks like only three submissions have passed the human expert cutoff to potentially be considered for early sharing.
\nThis competition is a doozy, but that's why it's worth doing. So excited to see what comes out of it all.
\nSean
", "rawMarkdown": "Hello Rhiju et al,\n\nAre you able to share the current status of the early sharing prize? I'm wondering if I can find the associated notebooks, but don't see them clearly flagged. I suspect I may be reading the public leaderboard wrong, but it looks like only three submissions have passed the human expert cutoff to potentially be considered for early sharing.\n\nThis competition is a doozy, but that's why it's worth doing. So excited to see what comes out of it all.\n\nSean", "votes": null }, { "id": "3192802", "postDate": "05/03/2025 11:40:31", "content": "For me, I currently have no plans to disclose my solution because I used private data from my graduate lab—I simply cannot share those sequences at this time, these private sequences and PDB I used for validation. More importantly, making my approach public would significantly hurt my chances of competing for the gold medal.
", "rawMarkdown": "For me, I currently have no plans to disclose my solution because I used private data from my graduate lab—I simply cannot share those sequences at this time, these private sequences and PDB I used for validation. More importantly, making my approach public would significantly hurt my chances of competing for the gold medal.", "votes": null }, { "id": "3192813", "postDate": "05/03/2025 11:58:17", "content": "I was wanting to check if my interpretation was correct that it's been a huge amount of work for these three teams to get up to that level. As a hobbyist I appreciate this is highly valued work for your studies and present/future career. Thanks for your hard work on this, maybe someday (other life circumstances allowing) I'll be able to join in at that level.
", "rawMarkdown": "I was wanting to check if my interpretation was correct that it's been a huge amount of work for these three teams to get up to that level. As a hobbyist I appreciate this is highly valued work for your studies and present/future career. Thanks for your hard work on this, maybe someday (other life circumstances allowing) I'll be able to join in at that level.", "votes": null }, { "id": "3192999", "postDate": "05/03/2025 17:03:42", "content": "Thanks for the question and for sharing your excitement. Indeed, the Early Sharing Prizes are still open -- even the vfold-beating entries have not claimed the prizes, perhaps because they do not satisfy the eligibility requirements. Go for it!
As the guy in LITERAL last place just let me say I appreciate your encouragement! I look forward to replacing my toy model with one that starts climbing the leader board.
", "rawMarkdown": "As the guy in LITERAL last place just let me say I appreciate your encouragement! I look forward to replacing my toy model with one that starts climbing the leader board.", "votes": null } ], "comments": [ { "id": 3192798, "author_name": "seanmcgee2025", "author_url": "", "post_date": "05/03/2025 11:34:46", "content": "Hello Rhiju et al,
\nAre you able to share the current status of the early sharing prize? I'm wondering if I can find the associated notebooks, but don't see them clearly flagged. I suspect I may be reading the public leaderboard wrong, but it looks like only three submissions have passed the human expert cutoff to potentially be considered for early sharing.
\nThis competition is a doozy, but that's why it's worth doing. So excited to see what comes out of it all.
\nSean
", "votes": null, "replies": [ { "id": 3192802, "author_name": "sweetyheehee", "author_url": "", "post_date": "05/03/2025 11:40:31", "content": "For me, I currently have no plans to disclose my solution because I used private data from my graduate lab—I simply cannot share those sequences at this time, these private sequences and PDB I used for validation. More importantly, making my approach public would significantly hurt my chances of competing for the gold medal.
", "votes": null, "replies": [ { "id": 3192813, "author_name": "seanmcgee2025", "author_url": "", "post_date": "05/03/2025 11:58:17", "content": "I was wanting to check if my interpretation was correct that it's been a huge amount of work for these three teams to get up to that level. As a hobbyist I appreciate this is highly valued work for your studies and present/future career. Thanks for your hard work on this, maybe someday (other life circumstances allowing) I'll be able to join in at that level.
", "votes": null, "replies": [] } ] }, { "id": 3192999, "author_name": "rhijudas", "author_url": "", "post_date": "05/03/2025 17:03:42", "content": "Thanks for the question and for sharing your excitement. Indeed, the Early Sharing Prizes are still open -- even the vfold-beating entries have not claimed the prizes, perhaps because they do not satisfy the eligibility requirements. Go for it!
As the guy in LITERAL last place just let me say I appreciate your encouragement! I look forward to replacing my toy model with one that starts climbing the leader board.
", "votes": null, "replies": [] } ] } ] } ], "raw_markdown_by_id": { "3187403": "The data refresh is now complete, and it is great to see that most of the previously entered notebooks were able to generate RNA 3D structure predictions after this stress test. Special thanks to @inversion for coordinating the data refresh!\n\nEven more exciting, many of the previously leading notebooks remain highly competitive after the refresh, close to or even exceeding the `vfold_human_expert` baseline!\n\n**The Early Sharing Prizes are now open.**\n\nIf you are one of the first two to beat the`vfold_human_expert` baseline and publish a well-documented notebook that respects `temporal_cutoff`, you get the Early Sharing Prize - and a very special recognition in the history of biology as the author of the first superhuman automated models for 3D RNA structure prediction.\n\nTo help us track submissions, it would be helpful to label your Notebook with “Early Sharing Prize” at the beginning of the Notebook title, and to make a post describing your solution herein Discussion. The full Eligibility conditions for the Early Sharing Prize are on our [competition overview page](https://www.kaggle.com/competitions/stanford-rna-3d-folding/overview).\n\nGood luck!\n\nFor the hosts,\nRhiju", "3192798": "Hello Rhiju et al,\n\nAre you able to share the current status of the early sharing prize? I'm wondering if I can find the associated notebooks, but don't see them clearly flagged. I suspect I may be reading the public leaderboard wrong, but it looks like only three submissions have passed the human expert cutoff to potentially be considered for early sharing.\n\nThis competition is a doozy, but that's why it's worth doing. So excited to see what comes out of it all.\n\nSean", "3192802": "For me, I currently have no plans to disclose my solution because I used private data from my graduate lab—I simply cannot share those sequences at this time, these private sequences and PDB I used for validation. More importantly, making my approach public would significantly hurt my chances of competing for the gold medal.", "3192813": "I was wanting to check if my interpretation was correct that it's been a huge amount of work for these three teams to get up to that level. As a hobbyist I appreciate this is highly valued work for your studies and present/future career. Thanks for your hard work on this, maybe someday (other life circumstances allowing) I'll be able to join in at that level.", "3192999": "Thanks for the question and for sharing your excitement. Indeed, the Early Sharing Prizes are still open -- even the `vfold`-beating entries have not claimed the prizes, perhaps because they do not satisfy the eligibility requirements. Go for it!", "3193023": "As the guy in LITERAL last place just let me say I appreciate your encouragement! I look forward to replacing my toy model with one that starts climbing the leader board." }, "source": "meta" }