A systematic evaluation of in-context learning in large language models for antibody characterization

Abstract Large language models can learn new tasks through in-context learning (ICL), yet this ability remains underexplored for biological sequence classification. We evaluate ICL across 20 large language models on three antibody tasks: species-origin, antibody specificity, and isotype class classification. Few-shot prompting improves over zero-shot performance, but matching the performance of protein language model classifiers requires sequence-similar demonstrations. Building on this observation, we introduce a sequence similarity-based strategy for ICL in antibody sequence classification, Sim-ICL. Using 32-shot prompting, Sim-ICL achieves competitive performance on two of three tasks. Its simplicity makes few-shot ICL promising for antibody characterization, especially for researchers with limited coding expertise.

Authors

Institutions

Publication Details

Journal
Genome biology
Published
2026-08-13
DOI
https://doi.org/10.1186/s13059-026-04240-6
Primary Topic
vaccines and immunoinformatics approaches
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

A systematic evaluation of in-context learning in large language models for antibody characterization

Sin-Hang Fung, Chenyang Hong, Jingying Zhou, Qin Cao et al.
Genome biology
vaccines and immunoinformatics approaches
article

A systematic evaluation of in-context learning in large language models for antibody characterization

Sin-Hang Fung, Chenyang Hong, Jingying Zhou, Qin Cao, Brian Shing‐Hei Wong, Stephen Kwok‐Wing Tsui, Kelly Yichen Li, Zhenghao Zhang, Miao Chen, Kevin Y. Yip, Ran Wang
article en

Abstract

Abstract Large language models can learn new tasks through in-context learning (ICL), yet this ability remains underexplored for biological sequence classification. We evaluate ICL across 20 large language models on three antibody tasks: species-origin, antibody specificity, and isotype class classification. Few-shot prompting improves over zero-shot performance, but matching the performance of protein language model classifiers requires sequence-similar demonstrations. Building on this observation, we introduce a sequence similarity-based strategy for ICL in antibody sequence classification, Sim-ICL. Using 32-shot prompting, Sim-ICL achieves competitive performance on two of three tasks. Its simplicity makes few-shot ICL promising for antibody characterization, especially for researchers with limited coding expertise.

Genome biology
Chinese University of Hong Kong (HK), Sanford Burnham Prebys Medical Discovery Institute (US), Discovery Institute (US), Chinese University of Hong Kong, Shenzhen (CN)
National Natural Science Foundation of China, Faculty of Medicine, Chinese University of Hong Kong
Quality Education
Openalex Percentile: Top 16%
vaccines and immunoinformatics approaches
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

A systematic evaluation of in-context learning in large language models for antibody characterization — Sin-Hang Fung, Chenyang Hong, et al. · Genome biology (2026) | TGRS Research Map | TGRS