Brian John Parker

RNAElectra is an RNA foundation model pretrained using replaced-token detection, showing strong performance across RNA regulatory tasks and supporting interpretability.

NucEL improves genomic representation learning by using ELECTRA-style pre-training with single-nucleotide tokenization, achieving state-of-the-art performance on multiple tasks.

CS-Fold improves RNA secondary structure prediction by integrating compensatory mutations from phylogenetic trees into a deep learning framework, achieving a 5% higher F1 score on cross-family datasets.

CRMnet is a deep learning model that predicts gene expression from regulatory sequences in yeast, achieving high accuracy and identifying transcription factor binding sites through model interpretation.

The paper quantifies gene expression regulation at multiple layers in response to ER stress in cervical cancer cells over time, identifying shared and stress-specific regulatory events.