ELISA: An Interpretable Hybrid Generative AI Agent for Expression-Grounded Discovery in Single-Cell Genomics

2026/03/12 by Omar Coser · 1 voice
Biochemistry, Genetics and Molecular Biology · Computer Science · #cs.AI #q-bio.GN

paper · pdf

arxiv published 2026/03/12 · arxiv created 2026/07/31 · arxiv updated 2026/07/31

Abstract

Translating single-cell RNA sequencing (scRNA-seq) data into mechanistic biological hypotheses remains a critical bottleneck, as agentic AI systems lack direct access to transcriptomic representations while expression foundation models remain opaque to natural language. Here we introduce ELISA (Embedding-Linked Interactive Single-cell Agent), an interpretable framework that unifies scGPT expression embeddings with BioBERT-based semantic retrieval and LLM-mediated interpretation for interactive single-cell discovery. An automatic query classifier routes inputs to gene marker scoring, semantic matching, or reciprocal rank fusion pipelines depending on whether the query is a gene signature, natural language concept, or mixture of both. Integrated analytical modules perform pathway activity scoringacross 60+ gene sets, ligand-receptor interaction prediction using 280+ curated pairs, condition-aware comparative analysis, and cell-type proportion estimation all operating directly on embedded data without access to the original count matrix. Benchmarked across six diverse scRNA-seq datasets spanning inflammatory lung disease, pediatric and adult cancers, organoid models, healthy tissue, and neurodevelopment, ELISA significantly outperforms CellWhisperer, a classical lexical retriever (BM25), and a random baseline in cell type retrieval (combined permutation test, p < 2×10^-5 for each), with particularly large gains on gene-signature queries (Cohen's d = 5.98 for MRR). ELISA replicates published biological findings (mean composite score 0.88), and generates candidate hypotheses through grounded LLM reasoning, bridging the gap between transcriptomic data exploration and biological discovery.

Citations

• Multimodal learning enables chat-based exploration of single-cell data
• The Virtual Lab of AI agents designs new SARS-CoV-2 nanobodies
• Biomni: A General-Purpose Biomedical AI Agent
• Longitudinal single-cell multiomic atlas of high-risk neuroblastoma reveals chemotherapy-induced tumor microenvironment rewiring
• The Llama 3 Herd of Models
• CellAgent: An LLM-driven Multi-Agent Framework for Automated Single-cell Data Analysis
• CRISPR-GPT for Agentic Automation of Gene-editing Experiments
• Empowering Biomedical Discovery with AI Agents
• Tokensome: Towards a Genetic Vision-Language GPT for Explainable and Cognitive Karyotyping
• Automated Bioinformatics Analysis via AutoBA
• GPT-4 Technical Report
• Large Language Models Encode Clinical Knowledge
• Large language models encode clinical knowledge
• Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks
• CellPhoneDB: inferring cell–cell communication from combined expression of multi-subunit ligand–receptor complexes
• Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks
• UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction

Discussions

• Omar Coser: ELISA: An Interpretable Hybrid Generative AI Agent for Expression-Grounded Discovery in Single-Cell Genomics https://arxiv.org/abs/2603.11872 https://arxiv.org/pdf/2603.11872 https://arxiv [bsky, 0 points, 0 comments]

Related

• DeepSeq: High-Throughput Single-Cell RNA Sequencing Data Labeling via Web Search-Augmented Agentic Generative AI Foundation Models
• A Multi-Modal AI Copilot for Single-Cell Analysis with Instruction Following
• SCTA: An Agentic Framework for Stable and Interpretable Target Gene Discovery from Single-Cell RNA Sequencing
• scBench: Evaluating AI Agents on Single-Cell RNA-seq Analysis
• Cell2Text: Multimodal LLM for Generating Single-Cell Descriptions from RNA-Seq Data