initialize_grids: For prediction, initialize grid points for cell cycle phase...

In peco: A Supervised Approach for **P**r**e**dicting **c**ell Cycle Pr**o**gression using scRNA-seq data

Description

For prediction, initialize grid points for cell cycle phase on a circle.

Usage

initialize_grids(Y, grids = 100, method.grid = c("pca", "uniform"))

Arguments

Y	Gene expression matrix. Gene by sample.
grids	number of bins to be selected along 0 to 2pi.
method.grid	The approach to initialize angles in the computation. uniform creates k equally-spaced bins (grids). pca uses gene expression values to infer angles, and then use these pca-based angles to move the cells to the closest bin (as defined by uniform).

Value

A vector of initialized angles to be used in cycle_npreg_loglik to infer angles.

Author(s)

Joyce Hsiao

See Also

cycle_npreg_loglik for log-likehood at angles between 0 to 2pi, cycle_npreg_mstep for estimating cyclic functions given inferred phases from cycle_npreg_loglik, cycle_npreg_outsample for predicting cell cycle phase using parameters learned from cycle_npreg_insample

Other peco classifier functions: cycle_npreg_insample, cycle_npreg_loglik, cycle_npreg_mstep, cycle_npreg_outsample

peco documentation built on Nov. 8, 2020, 8:16 p.m.