HiDeF: identifying persistent structures in multiscale 'omics data

In any 'omics study, the scale of analysis can dramatically affect the outcome. For instance, when clustering single-cell transcriptomes, is the analysis tuned to discover broad or specific cell types? Likewise, protein communities revealed from protein networks can vary widely in sizes depending on the method. Here, we use the concept of persistent homology, drawn from mathematical topology, to identify robust structures in data at all scales simultaneously. Application to mouse single-cell transcriptomes significantly expands the catalog of identified cell types, while analysis of SARS-COV-2 protein interactions suggests hijacking of WNT. The method, HiDeF, is available via Python and Cytoscape.

Automated summary

In this study, a new method called HiDeF is introduced, which utilizes the concept of persistent homology from mathematical topology to identify robust structures in data at all scales simultaneously. The application of HiDeF to mouse single-cell transcriptomes significantly expands the catalog of identified cell types, and the analysis of SARS-COV-2 protein interactions suggests hijacking of the WNT pathway. The HiDeF method is available through Python and Cytoscape.