In collaboration with colleagues from the Genome Institute of Singapore and the Bioinformatics Institute, members of the FER Laboratory for Bioinformatics and Computational Biology have published a paper in the prestigious journal Nature Communications. The paper presents, for the first time, the remarkable generalization capabilities of RNA language models across various biological problems and introduces the currently largest published RNA language model.

The title of the paper is "RiNALMo: General-Purpose RNA Language Models Can Generalize Well on Structure Prediction Tasks", and it is available here. The model implementation is publicly available in the corresponding GitHub repository.

The authors of the paper are Rafael Josip Penić, Tin Vlašić, Prof. Roland G. Huber, Prof. Yue Wan, and Prof. Mile Šikić.