Papers for

biotech researchers

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

Pretrained models improve search in complex discrete optimisation tasks

Generative Models as Adaptive Trust Regions for Discrete Bayesian Optimisation

Abstract: Bayesian optimisation (BO) is a powerful framework for sample-efficient optimisation of expensive black-box functions, but its application to high-dimensional discrete scientific design remains challenging. Local trust-region BO methods such as TuRBO address high dimensionality by restricting search to adaptively sized neighbourhoods. However, a direct discrete analogue based on edit distance can be poorly suited to structured discrete objects: neighbourhoods grow combinatorially, and metric proximity does not capture the dependencies that determine whether a design is plausible or useful. Motivated by the growing availability of pretrained generative models, we introduce GenTuRBO, which replaces the geometric trust region with an adaptively scaled proposal distribution defined by a pretrained generator. The generator provides a learned prior over discrete transformations, encoding coordinated changes without explicitly enumerating edit-distance neighbourhoods. GenTuRBO performs strongly across all three domains and achieves the highest mean final scores on Ehrlich sequence optimisation and Sokoban coordination.

Mon 12 OctMachine Learning
The gist
Optimising complex things made of discrete parts, like molecules or puzzles, is hard because the search space is huge and tricky to navigate. The authors replace a traditional method that searches nearby options by using pretrained generative models that better understand how parts change together. This approach, called GenTuRBO, adaptively focuses on promising changes learned from data instead of just counting simple edits. It works better than earlier methods on tasks like designing sequences and coordinating puzzle moves.
Open → 2610.16992v1

Discrete diffusion adaptation improves protein and DNA sequence design

D-DOIT: Training-free Adaptation of Discrete Diffusion via Doob's h-Transform

Abstract: We propose D-DOIT (Discrete Doob-Oriented Inference-time Transformation), a training-free and efficient adaptation method for discrete diffusion models with generic rewards. D-DOIT formulates adaptation as sampling from a reward-tilted target distribution and realizes this transport through Doob's h-transform of the discrete diffusion reverse kernel, using only reward values rather than reward gradients. Unlike continuous diffusion, masked discrete diffusion samples categorical token-reveal transitions rather than continuous state updates. D-DOIT derives the corresponding discrete Doob's h-transform, which guides sampling by reweighting reverse transition probabilities instead of adding a drift correction. To make this transformation practical, D-DOIT avoids expensive future rollouts. At each guided step, D-DOIT samples candidate next states, uses the model prediction head to complete each candidate into a clean sequence, evaluates each completion with the reward oracle, and resamples the next state with probabilities proportional to the rewards. An optional late-stage best-of-K refinement further improves sample quality by branching trajectories only near the end of denoising, avoiding the $K$-fold cost over the full trajectory. Empirically, across regulatory DNA design and protein inverse folding benchmarks, D-DOIT outperforms training-free guidance baselines. It improves enhancer activity and cell-type specificity while preserving sequence naturalness, and achieves the highest success rate in protein inverse folding.

Sun 4 OctMachine LearningArtificial Intelligence
The gist
Designing useful biological sequences like DNA or proteins is hard because many possible sequences exist. The authors present a new way to tweak existing models at sampling time without extra training, helping create sequences that score better on desired properties. Their method samples different sequence endings, evaluates them using rewards, and chooses better options, leading to more effective sequences. This approach works well on tasks like designing DNA that boosts cell activity and on folding proteins correctly.
Open → 2610.04938v1

Jepa-anything learns predictive models across diverse worlds and tasks

JEPA-Anything: Learning Predictive Models across Different Worlds

Abstract: World modeling enables intelligence to anticipate consequences, guide interventions, and learn from interaction. Yet predictive models remain domain-specific: can a common learning principle support world modeling across radically different systems? We introduce JEPA-Anything, a domain-agnostic framework based on orthogonal predictive factorization (OPF). Extending joint-embedding predictive architectures, OPF decomposes latent targets into complementary factors, learns them through dedicated pathways, and recombines them within a shared predictive design. We evaluate JEPA-Anything across seven domains: vision, biology, clinical trajectories, control, molecular dynamics, physical fields, and weather. Experiments span representation learning, intervention prediction, out-of-distribution generalization, and long-horizon dynamics, including 10 matched dynamics tasks, forecasting of over 1,000 clinical events, and 100-step molecular rollouts across four systems. Against matched JEPA baselines, JEPA-Anything improves reported metrics on all 10 dynamics tasks and reduces single-intervention prediction error on Interventional Pong by 34.8%. It achieves the lowest one-step and 100-step molecular errors among compared methods in all four systems. Beyond prediction, a factor-nominated biological intervention receives experimental support in cell co-cultures, patient-derived organoids, tumor fragments, and mice; latent orbital modes recover the Keplerian scaling exponent with a fitted slope of -1.4991. These results support a common factorized predictive principle across heterogeneous worlds, connecting world modeling with intervention and experimentally grounded scientific discovery. Code: https://github.com/Gen-Verse/JEPA-Anything

Thu 17 SeptComputation and Language
The gist
Making intelligent predictions about the future works well when focused on one kind of data, like images or physical movements. The authors introduce JEPA-Anything, a new method that can learn about many very different worlds using a shared approach. It breaks down predictions into parts, learns each, and combines them to improve accuracy across different fields like biology, weather, and molecular science. This shared method not only predicts better but can also help design new experiments and understand natural laws.
Open → 2609.20800v1