MARRVEL-MCP: A natural-language interface enabling LLMs to use 44 genomic tools for rare disease variant interpretation
By
Zhandong Liu2,3 Send email to [email protected]
Summary
Everton et al. introduce MARRVEL-MCP, a natural-language interface that allows large language models (LLMs) to access 44 genomic tools for rare disease variant interpretation. The system addresses usability barriers in variant interpretation tools like MARRVEL, which require precise formatting (e.g., HGVS notation) and return complex outputs. By enabling structured context engineering through tool-augmented reasoning, small models (3B–20B parameters) with tool access matched or outperformed larger models without tools, democratizing Mendelian disease discovery for non-experts and reducing cognitive burden on specialists.
Source
bskyMARRVEL-MCP: A natural-language interface enabling LLMs to use 44 genomic tools for rare disease variant interpretationcell.comKey quotes
· 3 pulledVariant interpretation in rare diseases requires navigating multiple genomic databases, each with strict input formats, while synthesizing heterogeneous evidence.
This process creates significant barriers for non-experts and imposes a substantial cognitive burden on experienced specialists.
With tool access, small models (3B–20B) matched or outperformed larger models without tools, highlighting structured context engineering in genomic reasoning.
You might also wanna read
AI-CURA: An LLM-based workflow achieves expert-level accuracy in genetic variant classification
This article presents a scientific study where researchers developed AI-CURA, an automated workflow using large language models (LLMs) — spe
scim.ag·5d ago
How LLMs Are Transforming Preclinical Drug Discovery Data Access
The article discusses how Large Language Models (LLMs) are being applied to preclinical drug discovery to help researchers efficiently acces
Reinforcement Learning Environment for LLMs in Cancer Diagnosis from Pathology Slides
A developer from Aluna (YC S24) presents a reinforcement learning environment for large language models focused on cancer diagnosis. The env
news.ycombinator.com·7mo ago
WebMCP: Enabling Websites to Share Tools and Resources with Client-Side LLMs
WebMCP is a proposal and implementation that enables websites to share tools, resources, and prompts with client-side large language models
github.com·8mo ago
Zebra-Llama: Efficient Hybrid Language Models Combining SSMs and Attention Layers
Researchers propose Zebra-Llama, a family of hybrid language models (1B, 3B, 8B) that combine State Space Models (SSMs) and Multi-head Laten
arxiv.org·6mo ago
MCP-Use: Open-Source Tool for Connecting LLMs to MCP Servers
MCP-Use is an open-source tool designed to connect any large language model (LLM) to MCP servers, enabling the creation of custom MCP agents
github.com·11mo ago
Comments
Sign in to join the conversation.
No comments yet. Be the first.