Paper List

期刊: ArXiv Preprint
发布日期: 2026-03-12
BioinformaticsAI/ML

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

No Affiliation

Omar Coser
Figure
Figure
Figure
Figure

30秒速读

IN SHORT: This paper addresses the critical bottleneck of translating high-dimensional single-cell transcriptomic data into interpretable biological hypotheses by bridging the gap between opaque expression foundation models and natural language interfaces.

核心创新

  • Methodology Introduces a hybrid retrieval architecture with automatic query classification that dynamically routes inputs to gene marker scoring, semantic matching, or reciprocal rank fusion pipelines based on query type.
  • Methodology Unifies scGPT expression embeddings with BioBERT semantic retrieval and LLM-mediated interpretation in a single interactive framework, enabling direct operation on embedded data without original count matrices.
  • Biology Develops integrated analytical modules for pathway activity scoring (60+ gene sets), ligand-receptor interaction prediction (280+ curated pairs), condition-aware comparative analysis, and cell-type proportion estimation.

主要结论

  • ELISA significantly outperforms CellWhisperer in cell type retrieval (combined permutation test, p<0.001) with particularly large gains on gene-signature queries (Cohen's d=5.98 for MRR).
  • The system replicates published biological findings with high fidelity (mean composite score 0.90) and near-perfect pathway alignment and theme coverage (0.98 each).
  • The hybrid retrieval architecture demonstrates complementary strengths: semantic pipeline excels on ontology queries while gene marker scoring dominates expression queries, with Union mode achieving optimal performance through adaptive routing.
研究空白： Existing AI systems either lack direct access to transcriptomic representations (agentic systems) or remain opaque to natural language interfaces (expression foundation models), creating a fundamental disconnect between data exploration and biological discovery.

摘要: 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 in cell type retrieval (combined permutation test, p<0.001), with particularly large gains on gene-signature queries (Cohen’s d=5.98 for MRR). ELISA replicates published biological findings (mean composite score 0.90) with near-perfect pathway alignment and theme coverage (0.98 each), and generates candidate hypotheses through grounded LLM reasoning, bridging the gap between transcriptomic data exploration and biological discovery. Code available at: https://github.com/omaruno/ELISA-An-AI-Agent-for-Expression-Grounded-Discovery-in-Single-Cell-Genomics.git (If you use ELISA in your research, please cite this work).

代码