findPCs: Find maximum recommended PC to use.

In weshorton/wrh.rUtils: Miscellaneous R Utility Functions

Find maximum recommended PC to use.

Description

Using a seurat object with PCA dimensional reduction run on it, determine the highest PC that still provides information about variance between cells. Adapted from: https://hbctraining.github.io/scRNA-seq/lessons/elbow_plot_metric.html

Usage

findPCs(seurat_obj, drName_v = "pca", min_v = T, elbow_v = T, ndims_v = 30)

Arguments

seurat_obj	A standard seurat object that has a "pca" dimensional reduction
drName_v	Name of pca dimensional reduction. Default is 'pca', which is Seurat default. If you create a PCA on specific genes and label it something else, provide that label name here (e.g. ccs_pca) for a PCA on cell cycle genes.
min_v	logical indicating to take the minimum cut-off of the two
elbow_v	logical indicating to make elbow plot
ndims_v	number of PCs to print in elbow plot

Details

Two cut-offs are calculated 1. The point where an individual PC contributes less than 5 and the cumulative PCs contribute at least 90 1. The point where the percent change in variation between the consecutive PCs is less than 0.1

Value

A single numeric value with the recommended PC cut-off

weshorton/wrh.rUtils documentation built on Oct. 28, 2024, 7:24 a.m.