My rotation project in the Church lab was on structure prediction for bacterial T-box riboswitches (the goal was to classify which tRNA was recognized by the riboswitch; fortunately we only needed to predict the secondary structure for this goal). Even these were rather difficult to model. We looked into deep learning methods but ended up going with a HMM approach due to lack of structural data. And of course, these are ribo switches, which change structure upon ligand binding.
So it's very cool to see the progress that has been made since 2020.
Author made it clear this was an educational essay, but concluding the problem has very limited therapeutic applications comes across like a bit of a take down for Atomic AI's platform.
I think it has very limited therapeutic applications with what we know about RNA structure today! But there's a great deal of completely unknown RNA biology (some of which I touch on in the essay) that may greatly benefit from RNA structure. The bit I mention about Arrakis Therapeutics preclinical work in drugging the (structured) RNA version of the MYC protein points to that being a very real possibility. All interesting biotech startups are built on bets on where the future is going, and I'm very happy that someone (AtomicAI and others) is betting on this, because clearly the answer of 'is RNA structure useful' isn't super open-and-shut
RNA structures are really more of a basic research thing. Having better tools there would be useful to understand these parts better. That's not irrelevant, but it doesn't lead directly to therapeutic applications.
RNAs so far have been very bad drug targets. That is to a large part inherent in their properties, they have fewer different components (4 bases compared to 20 amino acids) and the RNA backbone is strongly charged and interactions with something like that are generally unspecific. Odds are that RNA will remain a bad drug target for almost all cases.
I don’t disagree with your point, but I just would like to point out that there are over 100 known post-transcriptional modified RNA bases [1]. In fact, tRNA are more modified bases than canonical if taken as a whole. AND! the ribosome can’t function without all of its modifications. If I were to put money toward “targeting an RNA to make a drug” rRNA is where I’d aim…
the post is easy to read - and yet there's wealth of information together with citations of science papers info comes from
I love how author researched and discusses opinions opposite of his too
perhaps most surprising was to learn that vaccine RNA differs from normal U-A-C-G memorized in school. One more piece of knowledge I can come back with to my teachers one day :)
My rotation project in the Church lab was on structure prediction for bacterial T-box riboswitches (the goal was to classify which tRNA was recognized by the riboswitch; fortunately we only needed to predict the secondary structure for this goal). Even these were rather difficult to model. We looked into deep learning methods but ended up going with a HMM approach due to lack of structural data. And of course, these are ribo switches, which change structure upon ligand binding.
So it's very cool to see the progress that has been made since 2020.
OP, one small correction: the abbreviation of "long non-coding RNAs" is lncRNA (lowercase "L"), not "IncRNA".
Alphafold3 never claimed to solve rna structure prediction though
oh yeah, I didn’t mean to say that they did claim that, that was just my (mis)conception
Author made it clear this was an educational essay, but concluding the problem has very limited therapeutic applications comes across like a bit of a take down for Atomic AI's platform.
I think it has very limited therapeutic applications with what we know about RNA structure today! But there's a great deal of completely unknown RNA biology (some of which I touch on in the essay) that may greatly benefit from RNA structure. The bit I mention about Arrakis Therapeutics preclinical work in drugging the (structured) RNA version of the MYC protein points to that being a very real possibility. All interesting biotech startups are built on bets on where the future is going, and I'm very happy that someone (AtomicAI and others) is betting on this, because clearly the answer of 'is RNA structure useful' isn't super open-and-shut
RNA structures are really more of a basic research thing. Having better tools there would be useful to understand these parts better. That's not irrelevant, but it doesn't lead directly to therapeutic applications.
RNAs so far have been very bad drug targets. That is to a large part inherent in their properties, they have fewer different components (4 bases compared to 20 amino acids) and the RNA backbone is strongly charged and interactions with something like that are generally unspecific. Odds are that RNA will remain a bad drug target for almost all cases.
I don’t disagree with your point, but I just would like to point out that there are over 100 known post-transcriptional modified RNA bases [1]. In fact, tRNA are more modified bases than canonical if taken as a whole. AND! the ribosome can’t function without all of its modifications. If I were to put money toward “targeting an RNA to make a drug” rRNA is where I’d aim…
Source: PhD in RNA modifications
[1] https://pmc.ncbi.nlm.nih.gov/articles/PMC9073955/
wow, what an awesome biochemistry blog
the post is easy to read - and yet there's wealth of information together with citations of science papers info comes from
I love how author researched and discusses opinions opposite of his too
perhaps most surprising was to learn that vaccine RNA differs from normal U-A-C-G memorized in school. One more piece of knowledge I can come back with to my teachers one day :)