AI-CURA: An LLM-based workflow achieves expert-level accuracy in genetic variant classification
By
Brian Hon Yin Chung
Summary
This article presents a scientific study where researchers developed AI-CURA, an automated workflow using large language models (LLMs) — specifically DeepSeek-R1 — to classify genetic variants according to American College of Medical Genetics and Genomics (ACMG) guidelines. The system performed on par with human experts in classifying expert-curated genetic variants and enabled reclassification of variants that previously had uncertain classifications. The work addresses a key bottleneck in clinical genetics: the extensive literature analysis and human data interpretation required for genetic sequencing interpretation.
Source
Twitter / XAI-CURA: An LLM-based workflow achieves expert-level accuracy in genetic variant classificationscim.agKey quotes
· 3 pulledMa et al. developed a large language model (LLM)–based workflow to aid genetic interpretation according to American College of Medical Genetics and Genomics guidelines.
When incorporating the open-source LLM DeepSeek-R1, their pipeline performed on par with human experts at classifying expert-curated genetic variants
[The workflow] enabled reclassification of variants with uncertain classification
You might also wanna read
MARRVEL-MCP: A natural-language interface enabling LLMs to use 44 genomic tools for rare disease variant interpretation
Everton et al. introduce MARRVEL-MCP, a natural-language interface that allows large language models (LLMs) to access 44 genomic tools for r
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
UC San Diego AI Model Predicts Cancer Treatment Response from Tumor Genetics
UC San Diego researchers have developed an AI tool called iAtlas that can predict how a tumor will respond to treatment based solely on its
New Variant of DeepSeek AI Model Developed by German Lab TNG Technology Consulting GmbH
A new variant of the DeepSeek AI model, R1-0528, developed by German lab TNG Technology Consulting GmbH, is 200% faster than its predecessor
venturebeat.com·1y ago
Technical Implementation of DeepSeek LLM Deployment with Expert Parallelism on 96 H100 GPUs
The article details the technical implementation of deploying DeepSeek, an open-source large language model, across 96 H100 GPUs using advan
lmsys.org·10mo 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
Comments
Sign in to join the conversation.
No comments yet. Be the first.