DEtime_rank: Calculating the log-likelihood ratio of the biological time...

In ManchesterBioinference/DEtime: inferring the perturbation time from biological time course data

Calculating the log-likelihood ratio of the biological time course data

Description

DEtime_rank intends to rank biological time course data measured under control and perturbed conditions. In the function, an independent GP model and a combined GP model are used to model the data separately, the difference between the log-likelihood is used as the rank factor for the dataset. A high rank factor normally indicates better differential expression.

Usage

DEtime_rank(ControlTimes, ControlData, PerturbedTimes, PerturbedData,
  gene_ID = NULL, bound.lengthscale = NULL, savefile = TRUE)

Arguments

ControlTimes	Experimental time point at which time course biological data for the control case are measured, they have to be repeated if there are replicated measurements
ControlData	Time course data measured under control condition
PerturbedTimes	Experimental time point at which time course biological data for the perturbed case are measured, they have to be repeated if there are replicated measurements
PerturbedData	Time course data measured under perturbed condition
gene_ID	ID of these genes addressed in this study
bound.lengthscale	bounds for the lengthscale used in the DEtime RBF kernel. When not provided,bound.lengthscale <- c(max(diff(c(ControlTimes,PerturbedTimes))), 4*max(c(ControlTimes,PerturbedTimes)))
savefile	A BOOLEAN argument which is used to indicate if the ranking list will be saved or not

Details

ControlTimes and PerturbedTimes can be ordered by either time series, for instance time1, time1, time2, time2, time3, time3 ... or replicate sequences, for instance: time1, time2, time3, time1, time2, time3. ControlData and PerturbedData are two matrices where each row represents the time course data for one particular gene under either control or perturbed condition. The orders of the ControlData and PeruturbedData have to match those of the ControlTimes and PerturbedTimes, respectively.

Value

DEtime_rank returns a dataframe object whose first column is the gene_ID and second column is the Loglikelihood_ratio of the named gene.

Examples

### import simulated data
data(SimulatedData)
### calculating loglikelihood ratio
res <- DEtime_rank(ControlTimes = ControlTimes, ControlData = ControlData, 
       PerturbedTimes = PerturbedTimes,PerturbedData=PerturbedData)

ManchesterBioinference/DEtime documentation built on Feb. 9, 2024, 12:10 p.m.