LR_diff: Likelihood-based Tests for Detecting Differential Circadian...

In diffCircadian/diffCircadian: An R package for circadian and differential circadian analysis.

Likelihood-based Tests for Detecting Differential Circadian Pattern

Description

Likelihood-based tests for differential circadian pattern detection.

Usage

LR_diff(
  tt1,
  yy1,
  tt2,
  yy2,
  period = 24,
  method = "LR",
  FN = TRUE,
  type = "all"
)

Arguments

tt1	Time vector of condition 1.
yy1	Expression vector of condition 1.
tt2	Time vector of condition 2.
yy2	Expression vector of condition 2.
period	Period of the since curve. Default is 24.
method	Test used to detect differential circadian pattern. It can be chosen either "LR" or "Wald". Default is LR.
FN	"TRUE" if using finite sample likelihood-based tests and "FALSE" if using general large sample likelihood-based tests. Default is "TRUE".
type	Test differential circadian pattern in differential "amplitude", "phase", "basal", "fit" or "all". Default is "all".

Details

Test differential amplitude/phase/basal level/rhythmicity of circadian curve fitting using likelihood-based tests.

Value

A list, see details below. Formula 1: yy = amp * sin(2π/period * (phase + tt)) + offset. Formula 2: yy = A * sin(2π/period * tt) + B * cos(2*π/period * tt) + offset.

amp_1	Amplitude estimate of the 1st data.
amp_2	Amplitude estimate of the 2nd data.
amp_c	Amplitude estimate pooling all data together.
phase_1	Phase estimate of the 1st data, phase is restricted in (0, period).
phase_2	Phase estimate of the 2nd data, phase is restricted in (0, period).
phase_c	Phase estimate pooling all data together, phase is restricted in (0, period).
offset_1	Basal level estimate of the 1st data.
offset_2	Basal level estimate of the 2nd data.
offset_c	Basal level estimate pooling all data together.
sigma2_1	Variance estimate of the 1st data.
sigma2_2	Variance estimate of the 2nd data.
sigma2_C	Variance estimate pooling all data together.
l0	Log likelihood under the null (same variance between the two groups).
l1	Log likelihood under the alternative (different variance between the two groups).
stat	Test statistic.
pvalue	P-value from the test.

Author(s)

Zhiguang Huo, Haocheng Ding

Examples

set.seed(32608)
n <- 50
tt1 <- runif(n,0,24) 
Amp1 <- 2
Phase1 <- 6
Offset1 <- 3
yy1 <- Amp1 * sin(2*pi/24 * (tt1 + Phase1)) + Offset1 + rnorm(n,0,1)
tt2 <- runif(n,0,24) 
Amp2 <- 3
Phase2 <- 5
Offset2 <- 2
yy2 <- Amp2 * sin(2*pi/24 * (tt2 + Phase2)) + Offset2 + rnorm(n,0,1)
LR_diff(tt1, yy1, tt2, yy2)

diffCircadian/diffCircadian documentation built on Feb. 24, 2023, 9:07 a.m.