BAAI/OPI-Struc

OPI-Struc（开放蛋白质结构指令）是一个专门为STELLA项目设计的多模态指令调优数据集。它通过将蛋白质3D结构信息与文本指令相结合，扩展了OPI范式，使大型语言模型能够基于序列和结构模态执行蛋白质相关任务。OPI-Struc专注于两个关键的蛋白质相关任务：功能描述预测（FP）和酶催化反应预测（EP）。功能描述预测（FP）包括自由文本问答（FTQA）和多项选择问答（MCQA）格式，用于从蛋白质3D结构预测其生物学功能；酶催化反应预测（EP）用于从蛋白质3D结构预测酶名称或EC编号。总数据集大小：351,183个训练样本和40,993个测试样本。

OPI-Struc (Open Protein Instructions for Structures) is a multimodal instruction-tuning dataset specifically designed for the STELLA project. It extends the OPI paradigm by integrating protein 3D structure information with textual instructions, enabling LLMs to perform protein-related tasks grounded in both sequence and structural modalities. OPI-Struc focuses on two critical protein-related tasks: Functional Description Prediction (FP) and Enzyme-catalyzed Reaction Prediction (EP). Functional Description Prediction (FP) predicts the biological function of a protein from its 3D structure, in both free-text QA (FTQA) and multiple-choice QA (MCQA) formats. Enzyme-catalyzed Reaction Prediction (EP) predicts the enzyme name or EC number from a proteins 3D structure. Total dataset size: 351,183 training samples and 40,993 testing samples.