mixedcirc_integrate: Circadian rhythm integration

In PayamEmami/mixedcirc: Circadian rhythm analysis using mixed models

Circadian rhythm integration

Description

This functions performs circadian rhythm data integration using canonical correlations.

Usage

mixedcirc_integrate(
  data_input = NULL,
  time = NULL,
  group = NULL,
  id = NULL,
  period = 24,
  ncomp = 2,
  variables = NULL,
  lm_method = c("lm", "lme", "nlme")[2],
  f_test = c("multcomp_f", "multcomp_chi", "Satterthwaite", "Kenward-Roger")[3],
  abs_phase = TRUE,
  decompose = FALSE,
  center = TRUE,
  scale = FALSE,
  merge = FALSE,
  no_correlation = FALSE,
  no_interaction = FALSE,
  max.iter = max.iter,
  verbose = FALSE,
  ...
)

Arguments

data_input	A named list of numerical matrices or data.frames (N*P) where in the rows are samples (N) and the columns are variables (P). Each element of the list is a data set
group	A character vector of length N. If performing differential circadian rhythm analysis, group is a factor, showing grouping of the samples. Analysis of two groups is supported at this stage! See details!
id	A vector of length N showing identity of each *unique* sample. See details
period	Period of circadian rhythm. Default: 24
ncomp	Number of components. If set to 0, maximum number of possible components will be calculated (default: 2)
variables	A named list where in each element (one per data set) is a vector of length ncomp. Each element of the vector must be the number of variables retained on that particular component
decompose	decomposes the Within variation in the data set with respect to id (default: FALSE)
center	Centers the columns of each data set to zero
scale	scale each data set by dividing each column by its standard deviation
merge	If TRUE, the groups and the type will be merge in a single matrix for regression (default: FALSE)
no_correlation	If TRUE, zero covariance will be set between time and groups (default: FALSE)
no_interaction	If TRUE, no interaction between groups and time is assumed. (default FALSE)
max.iter	Maximum number of iteration of model (default: 100000)
verbose	Show information about different stages of the processes. Default FALSE
...	additionl arguments to the regression function

Details

This method is based on SGCCA analysis All the data sets will be correlated together with the linearized rhythm and the group The resulting partial or average scores can be used in mixedcirc_detect to perform differential circadian rhythm analysis using mixed models The loadings can be used for example to cluster the corresponding variables across different domains The scores can be retrieved using mixedcirc_getscore

Value

A class of mixedcirc_integration.

Examples

library(mixedcirc)
data("circa_data")
data_input<-list(a=circa_data$data_matrix[,1:3],b=circa_data$data_matrix[,3:7])
results<-mixedcirc_integrate(data_input,time = circa_data$time,group = circa_data$group,id = circa_data$id)
data_matrix<-mixedcirc_getscore(results,type = "partial",merge = T)
fitted_data<-mixedcirc_detect(data_matrix,time = circa_data$time,group = circa_data$group,id = circa_data$id)
plot(fitted_data[1])

PayamEmami/mixedcirc documentation built on Jan. 15, 2025, 5:36 p.m.