The first dataset is the preeclampsia dataset, in which 19 profiles of blood samples were collected from 9.7 million. The data are from 45 women throughout their pregnancies, which include 33 features measured per cell. The clinical outcome of the covariate for this dataset was cell gestational age, which ranged from 8 to 28 weeks. The dataset is applied in CytoCo-set evaluation. Here is the file structure that fits in CytoCo-set input format (symbol "X(____)X" defined as file name describe): data folder fcs_info.csv filenames_X(trials_number)X.json pree_fcs all X(sample_file_name)X.fcs test_labels_X(trials_number)X.csv train_labels_X(trials_number)X.csv marker.csv tripletlists_X(covariate)X_X(trials_number)X X(covariate)X_tripletlist_subpick_test_rffX(medianpooling_or_maxpooling)X_sameX(same_threhold_percange)X_diffX(diff_threhold_percange)X.txt X(covariate)X_tripletlist_subpick_trainval_rffX(medianpooling_or_maxpooling)X_sameX(same_threhold_percange)X_diffX(diff_threhold_percange)X.txt The dataset uploaded all the sample's raw features data (fcs file) and some file examples. fcs_info.csv records all sample's true labels and covariates.