AFD-Instruction Advances Antibody Design With Instructions
Researchers on arXiv (submitted Feb 4, 2026) present AFD-Instruction, the first large-scale instruction dataset pairing antibody sequences with functional annotations for understanding and design. The dataset contains antibody understanding (inferring function from sequences) and antibody design (de novo sequence generation under functional constraints) and shows that instruction-tuning general LLMs improves performance across diverse antibody tasks. This links sequences to textual function descriptions, aiding therapeutic modeling.
Key Points
- 1Introduces AFD-Instruction, a large-scale instruction dataset pairing antibody sequences with functional annotations
- 2Demonstrates improved task performance after instruction-tuning general LLMs on antibody understanding and design
- 3Enables natural-language-guided de novo antibody design, accelerating therapeutic discovery and modeling workflows
Scoring Rationale
Strong dataset novelty and actionable instruction-tuning results, limited by single-source preprint status and domain specificity.
Sources
Public references used for this report.
Practice with real Logistics & Shipping data
90 SQL & Python problems · 15 industry datasets
250 free problems · No credit card
See all Logistics & Shipping problems
