Skip to content

Power Analysis

Orr Ashenberg edited this page Nov 30, 2021 · 5 revisions

We developed a strategy for determining the number of cells, samples, and regions required to capture defined signals of interest related to tumor heterogeneity in single-cell and spatial cancer studies. Key questions include

For sc/sn RNA-Seq experiments:

  • How many cells should be profiled per sample?
  • How many samples (i.e. patients) should be profiled?

For spatio-molecular experiments:

  • How many fields of view should be assessed per sample?
  • How large should the field of view be?
  • How many samples (i.e., patients) should be profiled?

Power analysis strategies for both types of experiments are detailed in the HTAPP: sc/sn RNA-Seq and Spatial Power Analysis writeup document. Jupyter notebooks with code demonstrating power analysis for sc/sn RNA-Seq experiments for MBC and neuroblastoma are found here. The code for the spatial power analysis will be released soon in a bioRxiv preprint.