DBSOMA: A Machine Learning Method that Identifies Chemical Modulators of Transcriptional States Uncovers Effectors of Beta-Cell Maturation

There are currently few high-throughput ways to determine the aptness of applying the transcriptional readout of a biological perturbation to specific systems. Herein we use density analysis of transcriptional correlations to computationally predict whether a given perturbation readout is relevant to Stem Cell derived islet (SC-Islet) maturation. The approach, Denisty Based Self-Organizing Map Analysis (DBSOMA), first learns patterns of gene expression represented in scRNA-seq sets by clustering genes with the Self-Organizing-Map (SOM) algorithm. Perturbation expression profiles and other gene lists are then projected onto the SOM grid, where the degree of clustering is determined by the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm. Filtering based on the degree of structure and the degree of overlap with a desired state yields candidate perturbagens for use in that system. Here we applied DBSOMA to SC-Islet maturation and identified known and novel regulators of ß-cell maturation. Overall design: The DBSOMA tool was used to predict compounds from among L1000 transcriptional response signatures which may affect beta cell maturation in vitro. A subset of these predicted compounds were then incorporporated into the standard SC-islet differentiation protocol in planar cultures. After treatment through stage 5 and a week a of stage 6, whole islet cultures were submitted for bulk RNA sequencing by Admera Health.