Multi-agent AI enables evidence-based cell annotation in single-cell transcriptomics
(2025)- Abstract
- Cell type annotation remains a critical bottleneck, with current methods often inaccurate and requiring extensive manual validation, particularly in disease contexts. While large language models (LLMs) show promise, they can be unreliable due to hallucinations. We developed CyteType, a multi-agent framework that generates competing hypotheses grounded in full expression data and study context, validates against external databases, and iteratively self-evaluates. Comprehensive benchmarking demonstrates that CyteType substantially outperforms reference-based and LLM-based methods, with self-generated confidence scores reliably identifying trustworthy annotations. CyteType transforms cell type annotation from label assignment into... (More)
- Cell type annotation remains a critical bottleneck, with current methods often inaccurate and requiring extensive manual validation, particularly in disease contexts. While large language models (LLMs) show promise, they can be unreliable due to hallucinations. We developed CyteType, a multi-agent framework that generates competing hypotheses grounded in full expression data and study context, validates against external databases, and iteratively self-evaluates. Comprehensive benchmarking demonstrates that CyteType substantially outperforms reference-based and LLM-based methods, with self-generated confidence scores reliably identifying trustworthy annotations. CyteType transforms cell type annotation from label assignment into evidence-grounded biological discovery. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/4963b8c7-9cf4-4197-a113-a4bf76806e1b
- author
- Ahuja, Gautam
; Antill, Alex
LU
; Su, Yi
; Dall'Olio, Giovanni Marco
; Basnayake, Sukhitha
; Karlsson, Göran
LU
and Dhapola, Parashar
LU
- organization
- publishing date
- 2025-11-07
- type
- Working paper/Preprint
- publication status
- published
- subject
- publisher
- bioRxiv
- DOI
- 10.1101/2025.11.06.686964
- language
- English
- LU publication?
- yes
- id
- 4963b8c7-9cf4-4197-a113-a4bf76806e1b
- date added to LUP
- 2025-11-10 09:14:39
- date last changed
- 2025-11-10 13:39:08
@misc{4963b8c7-9cf4-4197-a113-a4bf76806e1b,
abstract = {{Cell type annotation remains a critical bottleneck, with current methods often inaccurate and requiring extensive manual validation, particularly in disease contexts. While large language models (LLMs) show promise, they can be unreliable due to hallucinations. We developed CyteType, a multi-agent framework that generates competing hypotheses grounded in full expression data and study context, validates against external databases, and iteratively self-evaluates. Comprehensive benchmarking demonstrates that CyteType substantially outperforms reference-based and LLM-based methods, with self-generated confidence scores reliably identifying trustworthy annotations. CyteType transforms cell type annotation from label assignment into evidence-grounded biological discovery.}},
author = {{Ahuja, Gautam and Antill, Alex and Su, Yi and Dall'Olio, Giovanni Marco and Basnayake, Sukhitha and Karlsson, Göran and Dhapola, Parashar}},
language = {{eng}},
month = {{11}},
note = {{Preprint}},
publisher = {{bioRxiv}},
title = {{Multi-agent AI enables evidence-based cell annotation in single-cell transcriptomics}},
url = {{http://dx.doi.org/10.1101/2025.11.06.686964}},
doi = {{10.1101/2025.11.06.686964}},
year = {{2025}},
}