Predicting glycan structure from tandem mass spectrometry via deep learning

Curated set of LC-MS/MS data from glycomics studies. Used for training and applying CandyCrunch, a deep learning model to predict glycan structure from LC-MS/MS data, described in Urban et al., bioRxiv, 2023 and https://github.com/BojarLab/CandyCrunch. Files: full_dataset.xlsx: Full dataset with all annotated LC-MS/MS glycan spectra X_train.pkl: spectra and metadata from our training set y_train.pkl: labels from our training set X_test.pkl: spectra and metadata from our independent test set y_test.pkl: labels from our independent test set

本数据集为糖组学研究相关的液相色谱-串联质谱（LC-MS/MS）精选整理数据集，用于训练与应用CandyCrunch模型——一款可从LC-MS/MS数据中预测聚糖结构的深度学习模型，相关研究详情见Urban等人2023年发表于bioRxiv的论文及开源仓库https://github.com/BojarLab/CandyCrunch。数据集包含以下文件：
full_dataset.xlsx：收录全部已注释LC-MS/MS聚糖质谱的完整数据集
X_train.pkl：训练集的质谱数据与元数据
y_train.pkl：训练集对应的标注标签
X_test.pkl：独立测试集的质谱数据与元数据
y_test.pkl：独立测试集对应的标注标签