Single-cell transcriptomics II

General information

Learning goals

• What?

  • Learn the definition and some popular methods for batch correction in scRNA-seq

  • Clustering scRNA-seq

  • Finding developmental trajectories of cells using popular scRNA-seq analysis methods

• Why?

  • It may be challenging to arrive to correct biological conclusions if the data are not normalized for the technical biases such as batch effect

  • Clustering can inform us about the presence of batch effects, cell subtypes, effects of cell perturbations, help in correcting cell type annotations

  • Using scRNA-seq one can understand how cells develop into their final type and how they acquire their functions

• How?

  • Batch correction methods: regression-based, dimensionality reduction-based, deep learningbased

  • Clustering methods: k-means-based, GMMs, hierarchical clustering-based, graph-based methods

  • Trajectory inference with PHATE and MONOCLE

Material

The slides for the lecture can be found here.