WeiYa's Work Yard

A dog, who fell into the ocean of statistics, tries to write down his ideas and notes to save himself.

scHOT: Investigate higher-order interactions in single-cell data

Posted on
Tags: Single-Cell, Pseudotime

This note is for Ghazanfar, Shila, Yingxin Lin, Xianbin Su, David Ming Lin, Ellis Patrick, Ze-Guang Han, John C. Marioni, and Jean Yee Hwa Yang. “Investigating Higher-Order Interactions in Single-Cell Data with scHOT.” Nature Methods 17, no. 8 (August 2020): 799–806.

high-order interactions (i.e., looking beyond changes in mean expression) have proved highly informative for understanding genomics data

a computationally derived prediction of cell-type differentiation trajectories, methods for identifying individual genes that substantially change their expression levels across the pseudotemporal trajectory typically focus on changes in mean expression of single genes and do not characterize subtle changes in patterns of covariation between subsets of genes across this trajectory.

scHOT identifies multiple higher-ordeer associations during liver development

when using Monocle 2 to order the cells in pseudotime, one observed a clear bifurcation where hepatoblasts differetiated into either cholangiocytes or hepatocytes.

  • first examine higher-order patterns as cell transitioned from naive hepatoblasts toward the bifurcation point where they commit to one of the two downstream lineages
  • next focus on the full trajectory from naive hepatoblast through to hepatocytes.
    • identified numerous changes in correlation between pairs of genes that did not change their individual mean expression
    • an example: Cdt1 and Top2a, which are protein-protein interacting partners that have been implicated in regulation of the cell cycle in human and mouse stem cells.
    • this pair of gene changes from being strongly negative correlated in the progenitor population to displaying no correlation in the more differentiated hepatocytes
    • but when consider them separately, neither Cdt1 nor Top2a are significantly differentially variable
  • explore whether the differential patterns of higher-order interactions could reveal genes that were related specifically to differentiation into the hepatocyte or cholangiocyte lineages, or if they reflected a common pattern of exit out of the hepatoblast state into mature cells.

Published in categories Note