Back

CIViC MCP: Integrating Large Language Models with the Clinical Interpretations of Variants in Cancer

Schimmelpfennig, L. E.; Cody, Q.; McMichael, J.; Coffman, A.; Saliba, J.; Danos, A.; Kiwala, S.; Wagner, A. H.; Sanz-Cruzado, J.; Lever, J.; Griffith, M.; Griffith, O. L.

2025-10-16 cancer biology
10.1101/2025.10.13.682185 bioRxiv
Show abstract

SummaryThe Clinical Interpretation of Variants in Cancer (CIViC) knowledgebase provides a community-driven, open-source platform for discussing the biological and Clinical Significance of molecular variants in cancer. To enable users to make complex connections between CIViC information, we developed the CIViC Model Context Protocol (MCP) server, which allows large language models (LLMs) to directly interface with the CIViC API through natural language, facilitating the rapid summarization of expertly curated cancer variant interpretations. Availability and implementationThe CIViC MCP server is detailed at https://github.com/griffithlab/civic-mcp-server with archived code and evaluation data deposited in Zenodo (DOI: 10.5281/zenodo.17344050). The repository is a fork of https://github.com/QuentinCody/civic-mcp-server (QuentinCody 2025) and includes instructions for accessing the server through the Claude desktop app and hosting it locally with GPT-5. We also provide a Python script for directly querying the MCP server. Supplemental informationSupplementary data are available at Bioinformatics online.

Published in Bioinformatics Advances (predicted rank #2) · training set

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

1
Bioinformatics
◐ 1204 papers in training set
Top 0.9%
22.6%
Similar papers in this journal
Bioinformatics Advances · published here
◐ 203 papers in training set
Top 0.2%
10.1%
Similar papers in this journal
3
Patterns
○ 78 papers in training set
Top 0.1%
10.1%
Similar papers in this journal
4
JCO Clinical Cancer Informatics
● 22 papers in training set
Top 0.1%
6.9%
Similar papers in this journal
5
PLOS Computational Biology
● 1863 papers in training set
Top 5%
6.9%
Similar papers in this journal
50% of probability mass above
6
Genome Biology
○ 637 papers in training set
Top 3%
3.6%
7
eLife
● 5828 papers in training set
Top 34%
3.3%
8
Database
◐ 61 papers in training set
Top 0.3%
2.5%
9
Cancer Research
● 130 papers in training set
Top 2%
1.8%
10
Nature Communications
○ 5641 papers in training set
Top 44%
1.8%
11
Nucleic Acids Research
◐ 1281 papers in training set
Top 9%
1.5%
12
Nature Methods
○ 385 papers in training set
Top 5%
1.5%
13
Genome Medicine
○ 183 papers in training set
Top 3%
1.4%
14
PLOS ONE
● 5266 papers in training set
Top 50%
1.4%
15
GigaScience
◐ 212 papers in training set
Top 3%
1.4%
16
iScience
○ 1154 papers in training set
Top 25%
1.2%
17
Genome Research
● 468 papers in training set
Top 5%
1.2%
18
npj Digital Medicine
○ 118 papers in training set
Top 3%
1.2%
19
PeerJ
◐ 308 papers in training set
Top 10%
1.0%
20
Journal of Translational Medicine
○ 57 papers in training set
Top 2%
0.9%
21
IEEE Transactions on Computational Biology and Bioinformatics
● 20 papers in training set
Top 0.6%
0.9%
22
Frontiers in Bioinformatics
○ 49 papers in training set
Top 2%
0.6%
23
BMC Bioinformatics
○ 457 papers in training set
Top 6%
0.6%

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.