PertMind: Eliciting Emergent Biological Reasoning in LLM via Reinforcement Learning on Cellular Perturbation Data

• URL: http://arxiv.org/abs/2608.16419v2
• Date: Sat, 22 Aug 2026 05:21:26 GMT
• Title: PertMind: Eliciting Emergent Biological Reasoning in LLM via Reinforcement Learning on Cellular Perturbation Data
• Abstract summary: We introduce PertMind, which combines trusted-trajectory supervised initialization with gene-, pathway-, and format-level reinforcement signals.<n>PertMind improved response inference in unseen cellular contexts while retaining general language capabilities.<n>It also transferred without task-specific post-training to reverse perturbation identification, double-perturbation reasoning, phenotypic-screen prioritization, and biological-process interpretation.
• Score: 19.246190225837665
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large language models can describe mechanisms, yet scalable post-training still depends on costly, manually curated biological reasoning traces. Here we show that cellular perturbation atlases can instead become reinforcement-learning environments, where measured gene responses provide computable rewards for biological reasoning. We introduce PertMind, which combines trusted-trajectory supervised initialization with gene-, pathway-, and format-level reinforcement signals. Trained only on forward perturbation-response prediction, PertMind improved response inference in unseen cellular contexts while retaining general language capabilities. It also transferred without task-specific post-training to reverse perturbation identification, double-perturbation reasoning, phenotypic-screen prioritization, and biological-process interpretation. PertMind further generated biological profiles that supported competitive gene, cell, and donor representations across multiscale downstream tasks. These results support the hypothesis that reinforcement on experimental endpoints can concentrate reusable biological strategies already accessible to pretrained models. More broadly, perturbation-derived reinforcement learning offers a scalable route for transforming expanding experimental atlases into training environments for general-purpose biological reasoning.

Related papers

• BioM-JEPA: joint-embedding prediction of graph-connected gene blocks in single cells [64.21963170814706]
  BioM-JEPA predicts aggregate representations of graph-connected gene blocks.<n>A student network infers each target-block representation from the remaining genes in a cell.<n>Across CellBench tasks, BioM-JEPA embeddings retained expression, pathway and neighbourhood information.
  arXiv  Detail & Related papers  (2026-08-06T11:58:29Z)
• PhageBench: Can LLMs Understand Raw Bacteriophage Genomes? [2.9367295285731285]
  Bacteriophages play a critical role in regulating microbial ecosystems and in antibiotic alternatives.<n>PhageBench is the first benchmark designed to evaluate phage genome understanding by mirroring the workflow of bioinformatics experts.
  arXiv  Detail & Related papers  (2026-04-07T12:14:23Z)
• Progressive Multi-Agent Reasoning for Biological Perturbation Prediction [32.71169480836875]
  We present LINCSQA, a novel benchmark for predicting target gene regulation under complex chemical perturbations.<n>We also propose PBio-Agent, a multi-agent framework that integrates difficulty-aware task sequencing with iterative knowledge refinement.<n>Our key insight is that genes affected by the same perturbation share causal structure, allowing confidently predicted genes to contextualize more challenging cases.
  arXiv  Detail & Related papers  (2026-02-07T06:59:44Z)
• SciHorizon-GENE: Benchmarking LLM for Life Sciences Inference from Gene Knowledge to Functional Understanding [30.790301729371475]
  Large language models (LLMs) have shown growing promise in biomedical research, particularly for knowledge-driven interpretation tasks.<n>We introduce SciHorizon-GENE, a large-scale gene-centric benchmark constructed from authoritative biological databases.<n>The benchmark integrates curated knowledge for over 190K human genes and comprises more than 540K questions covering diverse gene-to-function reasoning scenarios.
  arXiv  Detail & Related papers  (2026-01-19T08:06:35Z)
• BioReason: Incentivizing Multimodal Biological Reasoning within a DNA-LLM Model [12.528834366422466]
  BioReason learns to produce logical, biologically coherent deductions.<n>It boosts KEGG-based disease pathway prediction accuracy from 86% to 98%.<n>It also improves variant effect prediction by an average of 15% over strong baselines.
  arXiv  Detail & Related papers  (2025-05-29T15:49:27Z)
• TxPert: Leveraging Biochemical Relationships for Out-of-Distribution Transcriptomic Perturbation Prediction [11.083533122552396]
  We present TxPert, a new state-of-the-art method that leverages multiple biological knowledge networks to predict responses under OOD scenarios.<n>In particular, we present: (i) TxPert, a new state-of-the-art method that leverages multiple biological knowledge networks to predict responses under OOD scenarios; (ii) an in-depth analysis demonstrating the impact of graphs, model architecture, and data on performance; and (iii) an expanded benchmarking framework that strengthens evaluation standards for perturbation modeling.
  arXiv  Detail & Related papers  (2025-05-20T21:13:23Z)
• Contextualizing biological perturbation experiments through language [3.704686482174365]
  PerturbQA is a benchmark for structured reasoning over perturbation experiments.<n>We evaluate state-of-the-art machine learning and statistical approaches for modeling perturbations.<n>As a proof of feasibility, we introduce Summer (SUMMarize, retrievE, and answeR), a simple, domain-informed LLM framework.
  arXiv  Detail & Related papers  (2025-02-28T18:15:31Z)
• BioMaze: Benchmarking and Enhancing Large Language Models for Biological Pathway Reasoning [49.487327661584686]
  We introduce BioMaze, a dataset with 5.1K complex pathway problems from real research.<n>Our evaluation of methods such as CoT and graph-augmented reasoning, shows that LLMs struggle with pathway reasoning.<n>To address this, we propose PathSeeker, an LLM agent that enhances reasoning through interactive subgraph-based navigation.
  arXiv  Detail & Related papers  (2025-02-23T17:38:10Z)
• GENERator: A Long-Context Generative Genomic Foundation Model [66.46537421135996]
  We present GENERator, a generative genomic foundation model featuring a context length of 98k base pairs (bp) and 1.2B parameters.<n>Trained on an expansive dataset comprising 386B bp of DNA, the GENERator demonstrates state-of-the-art performance across both established and newly proposed benchmarks.<n>It also shows significant promise in sequence optimization, particularly through the prompt-responsive generation of enhancer sequences with specific activity profiles.
  arXiv  Detail & Related papers  (2025-02-11T05:39:49Z)
• BioDiscoveryAgent: An AI Agent for Designing Genetic Perturbation Experiments [112.25067497985447]
  We introduce BioDiscoveryAgent, an agent that designs new experiments, reasons about their outcomes, and efficiently navigates the hypothesis space to reach desired solutions.<n>BioDiscoveryAgent can uniquely design new experiments without the need to train a machine learning model.<n>It achieves an average of 21% improvement in predicting relevant genetic perturbations across six datasets.
  arXiv  Detail & Related papers  (2024-05-27T19:57:17Z)
• Causal machine learning for single-cell genomics [94.28105176231739]
  We discuss the application of machine learning techniques to single-cell genomics and their challenges. We first present the model that underlies most of current causal approaches to single-cell biology. We then identify open problems in the application of causal approaches to single-cell data.
  arXiv  Detail & Related papers  (2023-10-23T13:35:24Z)
• Contrastive-Signal-Dependent Plasticity: Self-Supervised Learning in Spiking Neural Circuits [61.94533459151743]
  This work addresses the challenge of designing neurobiologically-motivated schemes for adjusting the synapses of spiking networks. Our experimental simulations demonstrate a consistent advantage over other biologically-plausible approaches when training recurrent spiking networks.
  arXiv  Detail & Related papers  (2023-03-30T02:40:28Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.