AFD-INSTRUCTION: A Comprehensive Antibody Instruction Dataset with Functional Annotations for LLM-Based Understanding and Design
🤖 AI Summary
This work addresses the limitation of current large language models in antibody understanding and design, which stems from a lack of high-quality instruction data aligning antibody sequences with functional descriptions. To bridge this gap, the authors construct AFD-INSTRUCTION, the first large-scale instruction dataset specifically tailored for antibodies, encompassing two core tasks: antibody function comprehension and constraint-based generation. This dataset enables explicit alignment between antibody sequences and natural language functional annotations for the first time. Through instruction tuning on AFD-INSTRUCTION, general-purpose large language models achieve substantial performance gains across diverse antibody-related tasks. The study thus establishes a novel paradigm and provides a foundational resource for natural language–guided antibody modeling and therapeutic antibody discovery.
Application Category
📝 Abstract
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.
Problem
Research questions and friction points this paper is trying to address.
antibody
large language models
functional annotation
instruction dataset
sequence-function alignment
Innovation
Methods, ideas, or system contributions that make the work stand out.
antibody instruction tuning
sequence-function alignment
large language models
antibody design
functional annotation
🔎 Similar Papers
No similar papers found.
