spark: spark

In bernard2012/Giotto: Spatial single-cell transcriptomics toolbox

Description

Compute spatially expressed genes with SPARK method

Usage

spark(
  gobject,
  percentage = 0.1,
  min_count = 10,
  expression_values = "raw",
  num_core = 5,
  covariates = NULL,
  return_object = "data.table",
  ...
)

Arguments

gobject	giotto object
percentage	The percentage of cells that are expressed for analysis
min_count	minimum number of counts for a gene to be included
expression_values	type of values to use (raw by default)
num_core	number of cores to use
covariates	The covariates in experiments, i.e. confounding factors/batch effect. Column name of giotto cell metadata.
return_object	type of result to return (data.table or spark object)
...	Additional parameters to the spark.vc function

Details

This function is a wrapper for the method implemented in the SPARK package:

• 1. CreateSPARKObject create a SPARK object from a Giotto object

• 2. spark.vc Fits the count-based spatial model to estimate the parameters, see spark.vc for additional parameters

• 3. spark.test Testing multiple kernel matrices

Value

data.table with SPARK spatial genes results or the SPARK object

bernard2012/Giotto documentation built on Sept. 22, 2020, 10:29 a.m.