Assessing the Utility of Large Language Models for Phenotype-Driven Gene Prioritization in Rare Genetic Disorder Diagnosis

TL;DR

This study evaluates the performance of large language models in phenotype-driven gene prioritization for rare genetic disorder diagnosis, revealing their current limitations and potential for integration into genomic analysis workflows.

Contribution

It provides a comprehensive assessment of multiple LLMs in gene prioritization tasks, highlighting their strengths, limitations, and the impact of prompt complexity and model size.

Findings

GPT-4 achieved 16.0% accuracy, still below traditional tools.

Prediction accuracy improves with larger models.

Complex prompts increase task completeness but may reduce structure compliance.

Abstract

Phenotype-driven gene prioritization is a critical process in the diagnosis of rare genetic disorders for identifying and ranking potential disease-causing genes based on observed physical traits or phenotypes. While traditional approaches rely on curated knowledge graphs with phenotype-gene relations, recent advancements in large language models have opened doors to the potential of AI predictions through extensive training on diverse corpora and complex models. This study conducted a comprehensive evaluation of five large language models, including two Generative Pre-trained Transformers series, and three Llama2 series, assessing their performance across three key metrics: task completeness, gene prediction accuracy, and adherence to required output structures. Various experiments explored combinations of models, prompts, input types, and task difficulty levels. Our findings reveal that even the best-performing LLM, GPT-4, achieved an accuracy of 16.0%, which still lags behind traditional bioinformatics tools. Prediction accuracy increased with the parameter/model size. A similar increasing trend was observed for the task completion rate, with complicated prompts more likely to increase task completeness in models smaller than GPT-4. However, complicated prompts are more likely to decrease the structure compliance rate, but no prompt effects on GPT-4. Compared to HPO term-based input, LLM was also able to achieve better than random prediction accuracy by taking free-text input, but slightly lower than with the HPO input. Bias analysis showed that certain genes, such as MECP2, CDKL5, and SCN1A, are more likely to be top-ranked, potentially explaining the variances observed across different datasets. This study provides valuable insights into the integration of LLMs within genomic analysis, contributing to the ongoing discussion on the utilization of advanced LLMs in clinical workflows.