I've seen some discussion regarding what the targets (DMS and 2A3 reactivities), and my current understanding is this: a high reactivity means that that base is likely to be unpaired (thanks Pan for catching this). Since the ultimate goal of this competition is to predict the structure of an RNA molecule, however, is there a way to visualize, given just the reactivities for each base in an RNA sequence, the molecule's overall 3-D structure? I've seen visualization notebooks that do this, but they use other models (e.g., RhoFold, EternaFold) and not the reactivity models we're building here.
", "messages": [ { "id": "2534656", "postDate": "11/22/2023 19:23:56", "content": "I've seen some discussion regarding what the targets (DMS and 2A3 reactivities), and my current understanding is this: a high reactivity means that that base is likely to be unpaired (thanks Pan for catching this). Since the ultimate goal of this competition is to predict the structure of an RNA molecule, however, is there a way to visualize, given just the reactivities for each base in an RNA sequence, the molecule's overall 3-D structure? I've seen visualization notebooks that do this, but they use other models (e.g., RhoFold, EternaFold) and not the reactivity models we're building here.
", "rawMarkdown": "I've seen some discussion regarding what the targets (DMS and 2A3 reactivities), and my current understanding is this: a high reactivity means that that base is likely to be unpaired (thanks Pan for catching this). Since the ultimate goal of this competition is to predict the structure of an RNA molecule, however, is there a way to visualize, given just the reactivities for each base in an RNA sequence, the molecule's overall 3-D structure? I've seen visualization notebooks that do this, but they use other models (e.g., RhoFold, EternaFold) and not the reactivity models we're building here.", "votes": null }, { "id": "2534675", "postDate": "11/22/2023 19:42:50", "content": "A high reactivity actually means the base is very exposed to the chemical modification, and thus is likely to be unpaired.
\nNow to get to the 3D structure, there is still no clear path. It is more common to use the reactivities to predict the secondary structure, so which base is paired with which. You can use the reactivities as constraint for an algorithm like RNAStructure, or you can create a Mutate and Map profile
Thanks for your response! That actually makes a lot of sense. Thanks for catching my typo in the original statement as well.
", "rawMarkdown": "Thanks for your response! That actually makes a lot of sense. Thanks for catching my typo in the original statement as well.", "votes": null } ], "comments": [ { "id": 2534675, "author_name": "albericlajarte", "author_url": "", "post_date": "11/22/2023 19:42:50", "content": "A high reactivity actually means the base is very exposed to the chemical modification, and thus is likely to be unpaired.
\nNow to get to the 3D structure, there is still no clear path. It is more common to use the reactivities to predict the secondary structure, so which base is paired with which. You can use the reactivities as constraint for an algorithm like RNAStructure, or you can create a Mutate and Map profile
Thanks for your response! That actually makes a lot of sense. Thanks for catching my typo in the original statement as well.
", "votes": null, "replies": [] } ] } ], "raw_markdown_by_id": { "2534656": "I've seen some discussion regarding what the targets (DMS and 2A3 reactivities), and my current understanding is this: a high reactivity means that that base is likely to be unpaired (thanks Pan for catching this). Since the ultimate goal of this competition is to predict the structure of an RNA molecule, however, is there a way to visualize, given just the reactivities for each base in an RNA sequence, the molecule's overall 3-D structure? I've seen visualization notebooks that do this, but they use other models (e.g., RhoFold, EternaFold) and not the reactivity models we're building here.", "2534675": "A high reactivity actually means the base is very exposed to the chemical modification, and thus is likely to be unpaired. \nNow to get to the 3D structure, there is still no clear path. It is more common to use the reactivities to predict the secondary structure, so which base is paired with which. You can use the reactivities as constraint for an algorithm like RNAStructure, or you can create a Mutate and Map profile", "2534684": "Thanks for your response! That actually makes a lot of sense. Thanks for catching my typo in the original statement as well." }, "source": "meta" }