πDIA-CLIP: efficient identification of highly heterogeneous proteomics data via a generalized zero-shot framework

This dataset provides the data resources for testing and training πDIA-CLIP, including the full training dataset, the result files utilized for the analysis in the associated paper, and a curated demo subset. The demo MS files were generated by Bruderer R. et al. and are publicly available via ProteomeXchange with the dataset identifier PXD005573. For demonstration and workflow compatibility purposes, we have converted the original format into the following components: train_data.zip (25.13 GB): The full training dataset converted and prepared for model training. result_files.zip (10.43 GB): The analysis files containing the primary data, intermediate outputs, or final metrics used for the figures and tables presented in our paper. demo_data.zip (4.45 GB): A curated lightweight subset intended for users to quickly test the DIA-CLIP pipeline.