Zhiyuan Chen

BEACON introduces a comprehensive RNA benchmark to evaluate deep learning models, finding that single nucleotide tokenization and ALiBi improve performance.

CREaTor models cis-regulatory patterns using attention mechanisms and RNA-seq/ChIP-seq data, outperforming existing methods in predicting cCRE-gene interactions across cell types.

Uni-RNA, a deep learning model pre-trained on a large RNA sequence corpus, is reported to lead existing methods on structural and functional prediction tasks after fine-tuning.