Tutorials

scRNA

The following examples show the basic work flow for RNA datasets.

• Melanoma TME (Tirosh et al.)
  • 1. Load and clean data
  • 2 Typical SCANPY analysis
  • 3. SCMER Feature Selection
  • 4. Validation
  • 5. Session Info
• 198 Pan-Cancer Cell Lines Marker Selection
  • SCMER Feature Selection
  • Validation
• Ileum Lamina Propria Immunocyte Subtypes (Martin et al.)
  • Load data
  • SCANPY Analysis
  • SCMER Marker Selection
  • Validation
• Bone Marrow Immune Cell Differentiation
  • SCMER Feature Selection
  • Validation
• Dexamethasone Treated A549 Lung Cancer Cell Line (Cao et al.)
  • Data Preparation
  • SCANPY Analysis
  • SCMER Feature Selection
  • Validation

CyTOF

We also validated SCMER on CyTOF data, where there are far fewer features.

• Bone Marrow (Levine et al.)
  • Data Preparation
  • SCMER Marker Selection
  • Validation

CITE-Seq (Genes retaining protein manifold)

SCMER also has the ability to find genes that better represent the manifold defined by proteins. The genes encoding those proteins are not always the best.

• Multiomics Analysis: CBMC Cite-Seq (Feature Selection on Donor 1)
  • Process Protein
  • Process RNA
  • Compare Protein Profile with Gene Expression Profile
  • SCMER Feature Selection
  • Validation
• Multiomics Analysis: CBMC Cite-Seq (Validation on Donor 2)
  • Process Protein
  • Process RNA
  • Compare Protein Profile with Gene Expression Profile
  • Validation
• PBMC CITE-Seq Unbiased Feature Selection

Supplementary

These notebooks do not use SCMER. They support the validation of the analyses above. Other files such as gene sets supporting the analyses can be found here.

• Dexamethasone Treated A549 ATAC Peak Embedding
• PBMC CITE-Seq RNA Processing

Simulation

A simulation study is also available.

• Simulation
  • Metrics