scMEDAL: Interpretable Single-Cell Transcriptomics Analysis with Batch Effect Visualization via Deep Mixed-Effects Autoencoder

- AML_data.zip: Preprocessed AML (Acute Myeloid Leukemia) data and cross-validation splits. Original source: Gene Expression Omnibus (GEO), van Galen et al., 2019; accession GSE116256. - ASD_data.zip: Preprocessed ASD (Autism Spectrum Disorder) data and cross-validation splits. Original source: Velmeshev et al., 2019 — (https://autism.cells.ucsc.edu ). - HH_data.zip: Preprocessed Healthy Heart data and cross-validation splits. Original source: Figshare, Yu et al., 2023 (https://figshare.com/articles/dataset/Batch_Alignment_of_single-cell_transcriptomics_data_using_Deep_Metric_Learning/20499630/2) - outputs.zip: latent-space outputs from AE, AECct, scMEDAL-FE, scMEDAL-FEC, scMEDAL-RE, scVI, Harmony, Scanorama, SAUCIE, and scANVI, used to generate UMAP visualizations. It also contains the multibatch projection of 300-cell sample matrix for genomaps visualization, used to generate UMAP visualizations. - outputs_batchclassifier.zip: latent-space outputs from AECbatch, used to generate UMAP visualizations. Note: To reproduce the manuscript figures using the outputs data shared on this figshare, please refer to the analysis notebooks at: https://github.com/DeepLearningForPrecisionHealthLab/scMEDAL_for_scRNAseq.