GeneAgent: Self-verification Language Agent for Gene Set Knowledge Discovery using Domain Databases

• URL: http://arxiv.org/abs/2405.16205v1
• Date: Sat, 25 May 2024 12:35:15 GMT
• Title: GeneAgent: Self-verification Language Agent for Gene Set Knowledge Discovery using Domain Databases
• Authors: Zhizheng Wang, Qiao Jin, Chih-Hsuan Wei, Shubo Tian, Po-Ting Lai, Qingqing Zhu, Chi-Ping Day, Christina Ross, Zhiyong Lu,
• Abstract summary: We present GeneAgent, a first-of-its-kind language agent featuring self-verification capability. It autonomously interacts with various biological databases to improve accuracy and reduce hallucination occurrences. Benchmarking on 1,106 gene sets from different sources, GeneAgent consistently outperforms standard GPT-4 by a significant margin.
• Score: 5.831842925038342
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Gene set knowledge discovery is essential for advancing human functional genomics. Recent studies have shown promising performance by harnessing the power of Large Language Models (LLMs) on this task. Nonetheless, their results are subject to several limitations common in LLMs such as hallucinations. In response, we present GeneAgent, a first-of-its-kind language agent featuring self-verification capability. It autonomously interacts with various biological databases and leverages relevant domain knowledge to improve accuracy and reduce hallucination occurrences. Benchmarking on 1,106 gene sets from different sources, GeneAgent consistently outperforms standard GPT-4 by a significant margin. Moreover, a detailed manual review confirms the effectiveness of the self-verification module in minimizing hallucinations and generating more reliable analytical narratives. To demonstrate its practical utility, we apply GeneAgent to seven novel gene sets derived from mouse B2905 melanoma cell lines, with expert evaluations showing that GeneAgent offers novel insights into gene functions and subsequently expedites knowledge discovery.

Related papers

• GENERator: A Long-Context Generative Genomic Foundation Model [66.46537421135996]
  We present a generative genomic foundation model featuring a context length of 98k base pairs (bp) and 1.2B parameters. The model adheres to the central dogma of molecular biology, accurately generating protein-coding sequences. It also shows significant promise in sequence optimization, particularly through the prompt-responsive generation of promoter sequences.
  arXiv  Detail & Related papers  (2025-02-11T05:39:49Z)
• Survey and Improvement Strategies for Gene Prioritization with Large Language Models [61.24568051916653]
  Large language models (LLMs) have performed well in medical exams, but their effectiveness in diagnosing rare genetic diseases has not been assessed. We used multi-agent and Human Phenotype Ontology (HPO) classification to categorized patients based on phenotypes and solvability levels. At baseline, GPT-4 outperformed other LLMs, achieving near 30% accuracy in ranking causal genes correctly.
  arXiv  Detail & Related papers  (2025-01-30T23:03:03Z)
• GeneSUM: Large Language Model-based Gene Summary Extraction [20.181381276458488]
  We propose GeneSUM, a two-stage automated gene summary extractor utilizing a large language model (LLM) Our approach retrieves and eliminates redundancy of target gene literature and then fine-tunes the LLM to refine and streamline the summarization process.
  arXiv  Detail & Related papers  (2024-12-24T04:20:43Z)
• 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. BioDiscoveryAgent can uniquely design new experiments without the need to train a machine learning model. It achieves an average of 21% improvement in predicting relevant genetic perturbations across six datasets.
  arXiv  Detail & Related papers  (2024-05-27T19:57:17Z)
• VQDNA: Unleashing the Power of Vector Quantization for Multi-Species Genomic Sequence Modeling [60.91599380893732]
  VQDNA is a general-purpose framework that renovates genome tokenization from the perspective of genome vocabulary learning. By leveraging vector-quantized codebooks as learnable vocabulary, VQDNA can adaptively tokenize genomes into pattern-aware embeddings.
  arXiv  Detail & Related papers  (2024-05-13T20:15:03Z)
• Evaluation of large language models for discovery of gene set function [0.8864741602534821]
  We evaluate five Large Language Models (LLMs) for their ability to discover the common biological functions represented by a gene set. Benchmarking against canonical gene sets from the Gene Ontology, GPT-4 confidently recovered the curated name or a more general concept. In gene sets derived from 'omics data, GPT-4 identified novel functions not reported by classical functional enrichment.
  arXiv  Detail & Related papers  (2023-09-07T21:10:48Z)
• SemanticCAP: Chromatin Accessibility Prediction Enhanced by Features Learning from a Language Model [3.0643865202019698]
  We propose a new solution named SemanticCAP to identify accessible regions of the genome. It introduces a gene language model which models the context of gene sequences, thus being able to provide an effective representation of gene sequences. Compared with other systems under public benchmarks, our model proved to have better performance.
  arXiv  Detail & Related papers  (2022-04-05T11:47:58Z)
• Multi-modal Self-supervised Pre-training for Regulatory Genome Across Cell Types [75.65676405302105]
  We propose a simple yet effective approach for pre-training genome data in a multi-modal and self-supervised manner, which we call GeneBERT. We pre-train our model on the ATAC-seq dataset with 17 million genome sequences.
  arXiv  Detail & Related papers  (2021-10-11T12:48:44Z)
• Feature reduction for machine learning on molecular features: The GeneScore [58.720142291102135]
  The GeneScore is a concept of feature reduction for Machine Learning analysis of biomedical data. We show that the GeneScore is superior to a binary matrix in the classification of cancer entities.
  arXiv  Detail & Related papers  (2021-01-14T10:58:39Z)
• Mining Functionally Related Genes with Semi-Supervised Learning [0.0]
  We introduce a rich set of features and use them in conjunction with semisupervised learning approaches. The framework of learning with positive and unlabeled examples (LPU) is shown to be especially appropriate for mining functionally related genes.
  arXiv  Detail & Related papers  (2020-11-05T20:34:09Z)

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.