large language model

Lineage-mixed neurodevelopmental programs in pediatric sarcomas nominate tumor-selective surface antigen pairs.

This paper develops an AI framework to interpret single-cell RNA-seq data from pediatric solid tumors, identifying tumor-specific surface antigen pairs for targeted therapies.

Agentic Discovery Transfer Enables Cross-Cancer Discovery at Scale

SMART, a multi-agent framework, transfers findings from single-cell RNA-sequencing cancer studies to other cancer types, identifying cross-cancer prognostic signatures and generating new hypotheses.

From Closed-Set Annotation to In-Context Discovery: Biologically Grounded LLMs for Cellular Profiling

CellScribe is a generative framework that unifies cell identity and state by using a language model to generate cell type annotations and meta-program profiles, outperforming baselines and identifying new cell types through in-context learning.

Dataset structure outweighs method choice in single-cell cell-type annotation

In a 63-tool benchmark, dataset identity explained ≈84% of Cohen's κ variance on synthetic data versus ≈4% for tool identity, and k-nearest-neighbor (kNN) purity predicted accuracy near-linearly (R² = 0.85–0.99).

OmicsResonance: An LLM-assisted cloud ecosystem for interactiveand reproducible single-cell transcriptomic analysis

OmicsResonance provides a code-free web platform for scRNA-seq analysis, integrating LLM-assisted tools and interactive features to enable reproducible and transparent single-cell research.

SCALE: Scalable Conditional Atlas-Level Endpoint transport for virtual cell perturbation prediction

A set-aware latent transport model predicts treated cell populations from unpaired controls, keeps perturbation-specific signal in CRISPR, drug and cytokine data, and guided a cytokine choice tested in three donors.

BioPathfinder: Evidence-guided multi-agent platform enables hypothesis discovery for CAR-T engineering

A provenance-tracked multi-agent system curated 57 CAR-T patient scRNA-seq studies, generated reviewable hypotheses, and prioritized KLRC1/NKG2A, whose blockade improved CD8+ CAR-T antitumour activity in a Raji xenograft model.

Integrative AI-Enabled Virtual Cell Modeling Reveals a Clinically Relevant Latent Effector State of Human CD8 T Cells Undetectable by Conventional Analyses

The paper uses AI to model CD8 T-cell responses during ICI therapy, identifying a latent effector state that conventional methods miss.

VFB-MCP: Natural-Language Access to Drosophila Neuroscience Grounded by an Expert-Curated Ontology-Led Knowledgebase

Model Context Protocol tools over Virtual Fly Brain's graph let Claude answer 25/30 Drosophila neuroscience tasks precisely, vs 14/30 with web search, 2/30 bare.

OCellus: A Language-Model Framework for Single-Cell, Spatial, and Perturbation Biology with Natural-Language Reasoning

OCellus is a language model that addresses limitations in single-cell, spatial, and perturbation biology by integrating natural-language reasoning across multiple tasks.

CellTosg2Sequence: A Unified Text-Omics-Signaling-Graph Large Language Model for Single-Cell Analysis

CellTosg2Sequence integrates biomedical text and signaling graphs into a language model for single-cell analysis, improving performance over existing models.

scIsoAgent enables autonomous isoform-resolved characterization and sequence-informed interpretation of long-read single-cell transcriptomes

scIsoAgent improves long-read single-cell RNA-seq analysis by creating reproducible, isoform-resolved workflows and linking sequence data to functional hypotheses.

ConvergeCELL: An end-to-end platform from patient transcriptomics to therapeutic hypotheses

ConvergeCELL translates patient transcriptomics into therapeutic hypotheses by integrating patient representation, interpretability, and large language models, outperforming baselines in gene recovery and classification.

scConcept enables concept-level exploration of single-cell transcriptomic data

scConcept improves single-cell transcriptomic analysis by creating interpretable biological concepts from gene-level data, enhancing clustering and interpretability across multiple datasets.

Transcriptome-based cell type assignment for kidney cell culture models

The paper develops a transcriptome-based method to match bulk RNA-seq data from kidney cell lines to single-cell reference datasets, identifying accurate cell type identities and evaluating method performance.

CellVoyager: AI CompBio Agent Generates New Insights by Autonomously Analyzing Biological Data

CellVoyager autonomously analyzes scRNA-seq data to generate new hypotheses and findings, outperforming existing models and producing scientifically interesting results.

Phenotypic reversion and target prioritization for cellular inflammation via representation learning with foundation models

The paper uses single-cell foundation models to identify genetic targets that reverse inflammation in endothelial cells by analyzing Perturb-seq data with and without proinflammatory stimulation.

h5adify: neuro-symbolic metadata harmonizationenables scalable AnnData integration with locallarge language models

h5adify harmonizes metadata from single-cell datasets using neuro-symbolic methods and local LLMs, improving integration and reducing batch effects.

scChat: A Large Language Model-Powered Co-Pilot for Contextualized Single-Cell RNA Sequencing Analysis

scChat is a large language model co-pilot for scRNA-seq analysis that integrates contextual reasoning to improve hypothesis generation and experimental design.

scELMo: Embeddings from Language Models are Good Learners for Single-cell Data Analysis

scELMo uses Large Language Models to analyze single-cell data for clustering, batch correction, and cell-type annotation without retraining, outperforming existing methods.

An Agentic AI Framework for Ingestion and Standardization of Single-Cell RNA-seq Data Analysis

CellAtria automates scRNA-seq data ingestion and analysis through an AI agent and chatbot interface, reducing manual effort and increasing accessibility.

Generative single-cell transcriptomics via large language models

This paper introduces PGL, a large language model that generates synthetic single-cell RNA-seq data from metadata, enabling virtual cohort construction and reference-on-demand analysis.

scMOBA: A conversational single-cell Multi-Omics Brain Agent across species

scMOBA is a multi-omics brain agent that integrates single-cell and spatial data across species, achieving state-of-the-art cell type classification and improving downstream tasks like aging clock construction and disease prediction.

CellReasoner: A reasoning-enhanced large language model for cell type annotation

CellReasoner is a lightweight LLM for cell type annotation that uses reasoning to map gene expression to labels with zero- and few-shot generalization.

GPTAnno: Ontology-tree-guided hierarchical cell type annotation based on GPT models for single-cell data

GPTAnno provides a hierarchical cell type annotation method using GPT models and ontology trees, improving accuracy across species and tissues by integrating clustering with ontology constraints.

GRNITE: Gene Regulatory Network Inference with Text Embeddings

GRNITE improves gene regulatory network inference by combining text embeddings and graph neural networks, enhancing AUROC and recall across methods.

CellOntologyMapper: Consensus mapping of cell type annotation

CellOntologyMapper standardizes cell type annotations by mapping user-defined names to standardized ontological terms, improving data integration and analysis across diverse single-cell datasets.

A pre-trained large generative model for translating single-cell transcriptome to proteome

This paper introduces scTranslator, a pre-trained large generative model that translates single-cell transcriptomes to proteomes, showing accuracy and utility in multiple downstream tasks.

Tokenizing single-cell transcriptomes as a native language for large language models

This paper introduces CellTok, a method to convert single-cell transcriptomes into tokens for LLMs, enabling joint processing of cellular data and text.

Protein large language model assisted one-to-one gene homology mapping in cross-species single-cell transcriptome integration

The paper develops a protein large language model-based method to improve gene homology mapping in cross-species single-cell transcriptome integration, increasing homology pairs and identifying novel cell-type markers.