Gun Woo Byeon

Iterative deep learning-design of human enhancers exploits condensed sequence grammar to achieve cell type-specificity

This study uses iterative deep learning to design human enhancers with cell type-specific activity, validated through experimental testing and perturbation experiments.

Design of overlapping genes using deep generative models of protein sequences

The paper designs overlapping genes using deep generative models of protein sequences and shows high success rates in creating functional proteins in alternate reading frames.

Combinatorial optimization of mRNA structure, stability, and translation for RNA-based therapeutics

This study develops PERSIST-seq and In-line-seq to optimize mRNA stability and translation, finding that in-cell stability is more critical than ribosome load for protein output.

Functional and structural basis of extreme non-coding conservation in vertebrate 5'UTRs

The paper identifies conserved 5'UTRs as regulators of translation and reveals their structural dynamics using a new method.