Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2602.04916 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911701913305088 |
|---|---|
| author | Luo, Ling Jiang, Wenbin Chang, Hongyuan Wang, Xinkang Zhang, Xushi Xiong, Yueting Tong, Mengsha Yu, Rongshan |
| author_facet | Luo, Ling Jiang, Wenbin Chang, Hongyuan Wang, Xinkang Zhang, Xushi Xiong, Yueting Tong, Mengsha Yu, Rongshan |
| contents | Large language models (LLMs) have significantly advanced protein representation learning. However, their capacity to interpret and design antibodies through natural language remains limited. To address this challenge, we present AFD-Instruction, the first large-scale instruction dataset with functional annotations tailored to antibodies. This dataset encompasses two key components: antibody understanding, which infers functional attributes directly from sequences, and antibody design, which enables de novo sequence generation under functional constraints. These components provide explicit sequence-function alignment and support antibody design guided by natural language instructions. Extensive instruction-tuning experiments on general-purpose LLMs demonstrate that AFD-Instruction consistently improves performance across diverse antibody-related tasks. By linking antibody sequences with textual descriptions of function, AFD-Instruction establishes a new foundation for advancing antibody modeling and accelerating therapeutic discovery. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_04916 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | AFD-INSTRUCTION: A Comprehensive Antibody Instruction Dataset with Functional Annotations for LLM-Based Understanding and Design Luo, Ling Jiang, Wenbin Chang, Hongyuan Wang, Xinkang Zhang, Xushi Xiong, Yueting Tong, Mengsha Yu, Rongshan Quantitative Methods Computation and Language Large language models (LLMs) have significantly advanced protein representation learning. However, their capacity to interpret and design antibodies through natural language remains limited. To address this challenge, we present AFD-Instruction, the first large-scale instruction dataset with functional annotations tailored to antibodies. This dataset encompasses two key components: antibody understanding, which infers functional attributes directly from sequences, and antibody design, which enables de novo sequence generation under functional constraints. These components provide explicit sequence-function alignment and support antibody design guided by natural language instructions. Extensive instruction-tuning experiments on general-purpose LLMs demonstrate that AFD-Instruction consistently improves performance across diverse antibody-related tasks. By linking antibody sequences with textual descriptions of function, AFD-Instruction establishes a new foundation for advancing antibody modeling and accelerating therapeutic discovery. |
| title | AFD-INSTRUCTION: A Comprehensive Antibody Instruction Dataset with Functional Annotations for LLM-Based Understanding and Design |
| topic | Quantitative Methods Computation and Language |
| url | https://arxiv.org/abs/2602.04916 |