Locality-Controlled Out-of-Distribution Guidance for Robust Regulatory DNA Sequence Design
cs.LG, q-bio.GN
Submitted: 2026-10-30
Updated: 2026-10-30
Comments: Accepted to NeurIPS 2026
Code: https://github.com/Lotfollahi-lab/LocoGen
License: http://creativecommons.org/licenses/by-nc-sa/4.0/
The gist: Designing regulatory DNA sequences for targeted gene-expression control is an important challenge.
Terminology
Abstract
Designing regulatory DNA sequences for targeted gene-expression control is an important challenge. Recent approaches frame this task as property-conditioned generation, where diffusion models trained on natural sequences are guided toward high predicted activity through reward-based fine-tuning or sampling-time guidance. We show that single-reward evaluation can obscure reward-hacking behavior. By decomposing evaluation into property, novelty, and plausibility, we identify an empirically underexplored region of the evaluation space in which all three objectives are satisfied, but no existing baseline succeeds. We postulate that this gap arises because prior methods treat guidance primarily as a matter of magnitude. In regulatory DNA, however, sequence grammar is spatially nonuniform: motif positions are sensitive to perturbation, while background regions allow more exploration. We therefore propose LOCOGEN by reformulating guidance as a problem of locality. LOCOGEN learns an invariant motif mask through sequence-level invariance learning to protect motif positions, while applying OOD damping only to background positions. It operates at sampling time on a frozen diffusion backbone, without retraining the backbone. Across HepG2, K562, and SK-N-SH, LOCOGEN reaches this previously underexplored region of the property-novelty-plausibility space. Code is available at https://github.com/Lotfollahi-lab/LocoGen.
Sources
- Unsupervised Total Variation Loss for Semi-supervised Deep Learning of Semantic Segmentation
- Learning Sparse Neural Networks through $L_0$ Regularization
- Inference-Time Alignment in Diffusion Models with Reward-Guided Generation: Tutorial and Review
Related papers
- A Physics-Regulated Neural Framework for Learning 3D Grain Growth Dynamics
- FlexMoE: Learning One-for-All Multi-Budget Expert Pruning for Flexible MoE Deployment
- OptiMind: Improving Small Language Models for Optimization Modeling through Formulation Guidance and Multi-Turn Learning
- Staying in the Sweet Spot: Boost Reasoning Evolution via Responsive Rollout Difficulty Manipulation
- Examining the Impact of Data and Training Methods on Emergent and Subliminal Misalignment
- Rethinking Black-Box On-Policy Distillation: A Rubric-based Analysis