Ctrl-DNA applies constrained reinforcement learning to a genomic language model, generating promoter and enhancer sequences whose predicted activity is high in a target cell type and suppressed in off-target cell types.
Deciphering eukaryotic gene-regulatory logic with 100 million random promoters
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Ctrl-DNA: Controllable Cell-Type-Specific Regulatory DNA Design via Constrained RL
Ctrl-DNA applies constrained reinforcement learning to a genomic language model, generating promoter and enhancer sequences whose predicted activity is high in a target cell type and suppressed in off-target cell types.