lambda1_calculator {SparseMDC}

Lambda 1 Calcualtor

Description

Calculates the lambda 1 value for the SparseMDC algorithm. The lambda 1 value controls the number of marker genes selected for each cluster in the output from SparseMDC. It is calculated as the value of lambda 1 that results in no marker genes being selected when then are no meaningful clusters present in the data. Please see the original manuscript for further details.

Usage

lambda1_calculator(dat_l, dim, nclust, nboot = 1000, alpha1 = 0.5,
  delta = 1e-07)

Arguments

dat_l

list with D entries, each entry contains data d, p * n matrix. This data should be centered and log-transformed.

dim

Total number of conditions, D.

nclust

Total number of clusters.

nboot

The number of bootstrap repetitions used for estimating lambda 1, the default value is 1000.

alpha1

The quantile of the bootstrapped lambda 1 values to use, range is (0,1). The default value is 0.5, the median of the calculated lambda 1 values.

delta

Small value term added to ensure existance, default value is 0.0000001.

Value

The estimated value of lambda1 for use in main SparseMDC algorithm

Examples

set.seed(10)
# Select small dataset for example
data_test <- data_biase[1:100,]
# Split data into condition A and B
data_A <- data_test[ , which(condition_biase == "A")]
data_B <- data_test[ , which(condition_biase == "B")]
data_C <- data_test[ , which(condition_biase == "C")]
# Store data as list
dat_l <- list(data_A, data_B, data_C)
# Pre-process the data
pdat <- pre_proc_data(dat_l, dim=3, norm = FALSE, log = TRUE,
center = TRUE)
lambda1 <- lambda1_calculator(pdat, dim = 3, nclust = 3 )