RNA Structure Prediction Lags Behind Proteins as AlphaFold3 Struggles with Molecular Flexibility
Summary
RNA structure prediction remains far behind protein prediction as AlphaFold3 struggles with RNA's extreme flexibility, with only 7,759 RNA structures mapped versus 216,212 proteins, though most RNA therapeutics like mRNA vaccines rely on secondary structure which scientists can now predict with high accuracy.
Key Points
- RNA structure prediction remains significantly harder than protein structure prediction due to RNA's extreme flexibility and conformational diversity, with only 7,759 RNA structures in databases compared to 216,212 protein structures, and AlphaFold3 fails to fully solve the problem despite recent advances
- Most clinically relevant RNA therapeutics like mRNA vaccines, antisense oligonucleotides, and siRNAs depend primarily on secondary structure rather than complex tertiary structure, with secondary structure prediction now reaching high accuracy levels similar to protein structure prediction
- Tertiary RNA structure prediction shows limited therapeutic value except for specific applications like aptamers, ribozymes, and ribosome-targeting antibiotics, though emerging areas like circular RNAs and modified nucleotides in vaccines present new challenges requiring better structural modeling