SQUARNA - an RNA secondary structure prediction method based on a greedy stem formation model
Bohdan, D. R.; Nikolaev, G. I.; Bujnicki, J. M.; Baulin, E. F.
Show abstract
Non-coding RNAs play a diverse range of roles in various cellular processes, with their spatial structure being pivotal to their function. The RNAs secondary structure is a key determinant of its overall fold. Given the scarcity of experimentally determined RNA 3D structures, understanding the secondary structure is vital for discerning the molecules function. Currently, there is no universally effective solution for de novo RNA secondary structure prediction. Existing methods are becoming increasingly complex without marked improvements in accuracy, and they often overlook critical elements such as pseudoknots. In this work, we introduce SQUARNA, a novel approach to de novo RNA secondary structure prediction. This method utilizes a simple, greedy stem formation model, addressing many of the limitations inherent in previous tools. Our benchmarks demonstrate that SQUARNA matches the performance of leading methods for single sequence inputs and significantly surpasses existing tools when applied to sequence alignment inputs.
Matching journals
The top 5 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- A comprehensive survey of long-range tertiary interactions and motifs in non-coding RNA structures 98%
- ARTEMIS - a method for topology-independent superposition of RNA 3D structures and structure-based sequence alignment 98%
- UFold: Fast and Accurate RNA Secondary Structure Prediction with Deep Learning 96%
Similar papers in this journal
- RNA structure prediction using positive and negative evolutionary information 96%
- Structural comparison of protein-RNA homologous interfaces reveals widespread overall conservation contrasted with versatility in polar contacts 96%
- RNA covariation at helix-level resolution for the identification of evolutionarily conserved RNA structure 96%
Similar papers in this journal
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.