| metaAnalysis {MetaCyto} | R Documentation |
A function that performs meta-analysis
metaAnalysis(value, variableOfInterst, otherVariables, studyID, data, CILevel, main, ifScale = c(TRUE, FALSE), cex = 1)
value |
A string to specify the column name of the dependent variable (y) |
variableOfInterst |
A string to specify the column name of the independent variable of interest (x1) |
otherVariables |
A string vector to specify the column names of independent variables included in the regression model other than the variableOfInterst. |
studyID |
A string to specify the column name of study ID. |
data |
A data frame containing the data |
CILevel |
A number between 0 to 1, used to specify the confidence interval to be plotted in the forest plot. |
main |
A string to specify the title of the forest plot |
ifScale |
A vector of two logic values, specifying if the dependent variable and the variableOfInterst should be scaled when calculating the effect size. |
cex |
A number specifying the amount by which plotting text and symbols should be scaled relative to the default in the forest plot. |
Returns data frame describing the effect size of variableOfInterst on value in each individual studies, as well as the over all effect size.
library(dplyr)
#collect all summary statistics
fn=system.file("extdata","",package="MetaCyto")
files=list.files(fn,pattern="cluster_stats_in_each_sample",recursive=TRUE,
full.names=TRUE)
fcs_stats=collectData(files,longform=TRUE)
# Collect sample information
files=list.files(fn,pattern="sample_info",recursive=TRUE,full.names=TRUE)
sample_info=collectData(files,longform=FALSE)
# join the cluster summary statistics with sample information
all_data=inner_join(fcs_stats,sample_info,by="fcs_files")
# plot forrest plot to see if the proportion of CCR7+ CD8 T cell
# is affected by age (while controlling for Gender)
L="CD3+|CD4-|CD8+|CCR7+"
dat=subset(all_data,all_data$parameter_name=="fraction"&
all_data$label==L)
MA=metaAnalysis(value="value",variableOfInterst="Subject Age",main=L,
otherVariables=c("Gender"),studyID="study_id",
data=dat,CILevel=0.95,ifScale=c(TRUE,FALSE))