Sceptic: pseudotime analysis for time-series single-cell sequencing and imaging data

Image data generated for this project is available at the 4D Nucleome Consortium data portal (\url{https://data.4dnucleome.org/li_sceptic_mESC_microscope}) with accession IDs: embryonic stem cells (day 0): 4DNESGFHRFGV; Embryoid body day 3: 4DNESIIXJOYP; Embryoid body day 7: 4DNESHZJ83A4; Embryoid body day 11: 4DNESR8N1O2P; NPCs: 4DNESLP3I9VG. Accordingly, we collected F121-6 mESC differentiation 3D images (DNA staining with DAPI and immunostaining of nucleolin) using five time points. We segmented each 2D-maximum-projection image using a pre-trained Mask R-CNN model, and we used a pre-trained ResNet34 model to generate 1024-dimensional representations for each nucleus. To measure the performance of the classifier, we randomly selected 1,200 single-nucleus image crops from each time point, yielding 6,000 crops in total.