tathadn/biolite-interpret-data

BioLite-Interpret Data (v2) 是一个用于训练语言模型的数据集，旨在生成对差异表达表、GO/KEGG富集结果和综合分析结果的生物学解释。数据集包含1,262个示例，分为训练集、验证集和测试集，比例为85%/5%/10%。数据来源于合成生成和经过筛选的公开数据集（如Mol-Instructions、BioInstruct和GEO数据集对）。数据集按任务类型和数据来源进行了分层，以确保平衡性。此外，README还详细介绍了版本历史、数据来源、过滤方法、任务类型、示例格式、合成生成细节、分割方法以及预期用途。

BioLite-Interpret Data (v2) is a dataset designed for training language models to generate biological interpretations of differential expression tables, GO/KEGG enrichment results, and combined analyses. The dataset includes 1,262 examples, split into training, validation, and test sets with a ratio of 85%/5%/10%. The data comes from synthetic generation and filtered sources like Mol-Instructions, BioInstruct, and GEO dataset pairs. The dataset is stratified by task type and source, ensuring balanced representation. The README also details the version history, sources, filtering methodology, task types, example format, synthetic generation details, split methodology, and intended use.