Length matters—secondary structures of long, diverse RNAs in RNAndria and deep learning model eFold improve prediction accuracy.

The results of the Fifth RNA-Puzzles contest highlights advances in RNA three-dimensional structure prediction and uncovers new insights into RNA folding and structure.

The development of CPICANN, a novel convolutional self-attention neural network, represents a groundbreaking approach in materials informatics. By leveraging the convolutional self-attention mechanism, CPICANN automates and significantly enhances the efficiency of crystal phase identification from whole X-ray powder diffraction patterns, marking a substantial advancement over traditional time-consuming methods.

The database of 200 million protein-structure predictions now includes homodimers, adding new biological relevance.

This paper proposes a rigorous definition of a periodic structure as an equivalence class of periodic sets under rigid motion, which is a composition of translations and rotations.

TransMagNet is a branch network model that predicts crystal systems and space groups directly from chemical compositions with high accuracy, providing a novel approach for materials design and discovery without relying on structural data.

We provide all the PDB files that we analyzed for the article "Challenges in predicting PROTAC-mediated Protein-Protein Interfaces with AlphaFold" Change from version 2: removed case 2wdt from the ligand-mediated cases.