To further demonstrate the flexibility of our approach, we used bridge integration to map and annotate a snmC-seq dataset, which measures DNA methylation profiles in single cells from the human cortex52. As a reference, we utilized a dataset from the Allen Brain Atlas which defines a taxonomy of cell-types in the human cortex, and is accompanied by an expertly curated and multi-level cell ontology53. Using a snmC2T-seq dataset which simultaneously measures methylation and gene expression as a bridge28, we were able to annotate the snmC-seq profiles with high confidence (Supplementary Fig. 3f). Even when our reference-derived annotations did not augment the resolution to unsupervised clustering of snmC-seq data, they did add substantial interpretability (Fig. 3d-f). For example, unsupervised clustering identified multiple populations of L6 neurons (labeled as L6-1, L6-2, and L6-3), but RNA-assisted annotation clearly labeled these clusters as either ‘Near Projecting’ (NP) or deep neocortical laminar 6b (L6b) excitatory neurons (Fig. 3f).
