Inferring disease progression stages in single-cell transcriptomics using a weakly supervised deep learning approach.

Application of single-cell/nucleus genomic sequencing to patient-derived tissues offers potential solutions to delineate disease mechanisms in humans. However, individual cells in patient-derived tissues are in different pathological stages, and hence, such cellular variability impedes subsequent differential gene expression analyses. To overcome such a heterogeneity issue, we present a novel deep learning approach, scIDST, that infers disease progression levels of individual cells with weak supervision framework. The disease progression-inferred cells display significant differential expression of disease-relevant genes, which cannot be detected by comparative analysis between patients and healthy donors. In addition, we demonstrate that pretrained models by scIDST are applicable to multiple independent data resources and are advantageous to infer cells related to certain disease risks and comorbidities. Taken together, scIDST offers a new strategy of single-cell sequencing analysis to identify bona fide disease-associated molecular features.