MicrobTiSDA (version 0.1.0)

Pred.data.MESR: Predict Time-Series Data from Fitted Mixed-Effect Spline regression Models

Description

This function generates predicted OTU abundances at new time points using previously fitted spline regression models. For each group defined in the fitted models, it extracts the time range from the metadata, creates a new sequence of time points with a specified interval, and uses the corresponding fitted models to predict values at these new time points.

Usage

Pred.data.MESR(Fitted_models, metadata, Group, time_step, Sample_Time)

Value

An object of PredictedDataMESR. Each element of the list corresponds to a group and contains the predicted OTU abundances at the new time points, with rows labeled by the group and time (formatted as Group_T_Time) and an additional column Predicted_Time.

Arguments

Fitted_models

A list object generated by Reg.MESR.

metadata

A data frame. Containing information about all samples, including at least the grouping of all samples as well as individual information (Group and ID), the sampling Time point for each sample, and other relevant information.

Group

A character string indicating the column name in metadata that defines group information for prediction.

time_step

A numeric value specifying the interval between new time points in the prediction sequence.

Sample_Time

A character string indicating the column name in metadata that contains sample time information.

Author

Shijia Li

Details

The function accepts a list of fitted models (typically the output from Reg.MESR) and sample metadata. For each group (as defined by the Group parameter), it extracts the subset of metadata corresponding to that group. The minimum and maximum time values are determined, and a new sequence of time points is generated using the provided time_step. For each OTU model of the group, predictions are obtained via the predict function applied to the new time points, and the results are compiled into a data frame with an additional column Predicted_Time indicating the prediction time. The output is a list of data frames, each corresponding to a group, containing the predicted OTU abundances.

Examples

# \donttest{
metadata <- data.frame(
  TimePoint = c(1, 2, 3, 4),
  Sample = c('S1', 'S2', 'S3', 'S4'),
  GroupA = c('A', 'A', 'B', 'B'),
  GroupB = c('X', 'Y', 'X', 'Y')
)

# Example pre-processed data (e.g., transformed abundance data)
Pre_processed_Data <- data.frame(
  Feature1 = rnorm(4),
  Feature2 = rnorm(4)
)

# Create design matrix using grouping variables
design_data <- Design(metadata, Group_var = c('GroupA', 'GroupB'), Pre_processed_Data,
                      Sample_Time = 'TimePoint', Sample_ID = 'Sample')

fit_result <- Reg.MESR(Data_for_Reg = design_data,
                       pre_processed_data = Pre_processed_Data,
                       unique_values = 5,
                       z_score = 2,
                       Knots = NULL,
                       max_Knots = 5)

predictions <- Pred.data.MESR(fit_result,
                             metadata,
                             Group = "Group",
                             time_step = 1,
                             Sample_Time = "TimePoint")
# }