Mining single-cell data for cell type-disease associations

Supplementary Data, including code and tables, used in the manuscript.

Supplementary Figures

• S1-S4: Time course plots for each dataset
• S5-S8: Drug target classification for selected HPO terms in each dataset

Code

• Input processing for each dataset
• Generic-CountsToSeurat.R - code for applying sctransform framework to generate Seurat object from counts matrices and metadata
• EWCE
  • EWCE1-PrepareEWCEInputs.R - code for producing intermediate outputs from Seurat object
  • EWCE2-PerformEWCEEnrichments.R - code for using intermediate outputs to generate enrichments via EWCE
• hdWGCNA (Gene co-expression analysis)
  • hdWGCNA1-CountsToSeurat.R - code for preparing counts matrices and metadata into Seurat object for use in hdWGCNA
  • hdWGCNA2-SeuratTohdWGCNA.R - code for generating co-expression modules from input Seurat object
  • hdWGCNA3-hdWGCNAEnrichments.R - code for enriching co-expression modules for HPO terms
  • hdWGCNA4-ModuleCellTypeAssociations.R - code for associating co-expression modules with cell types
• TCseq (Temporal clustering analysis)
  • TCseq1-SeuratToTimeClusters.R - code for constructing temporal clusters from input Seurat object, and performing HPO enrichments on the clusters 
  • TCseq2-CombineEnrichments.R - code for collating enrichment results into a single csv file
• Drug target analysis
  • DrugTargets1-GetOpenTargetsData.py - code for extracting target/disease associations from OpenTargets data
  • DrugTargets2-CoexpressionModule_DrugTarget_Overlap.R - code for determining the distribution of drug targets across co-expression modules 
  • DrugTargets3- code for determining if drug targets were found in co-expression modules, as well as if they were in the HPO gene list 

Supplementary Tables

Table descriptions are included in the Excel file.