Davyd R. Bohdan

SQUARNA improves RNA secondary structure prediction by combining stem maximization with free energy minimization, outperforming existing methods including deep learning models.

ARTEM identifies RNA and DNA tertiary motifs by searching structural databases, revealing new kink-turn variants and their functional implications.

ARTEMIS improves RNA 3D structure superposition and alignment, revealing conserved helical motifs across non-coding RNAs with sequence permutations.

SQUARNA is an RNA secondary structure prediction method that uses a greedy stem formation model, performing competitively on single sequences and outperforming existing tools on sequence alignment inputs.

The paper surveys long-range tertiary interactions and motifs in non-coding RNA structures, identifies three common interaction types, and introduces a new dataset and algorithm for their analysis.