getDrift: Mean of residuals from drifting CpGs.

In metamaden/cgageR:

Description

Calculate mean of residuals from user-specified drifting CpGs.

Usage

getDrift(df, driftcg,chrage,showStatusDrift=TRUE,keepres=TRUE)

Arguments

df	Beta-value data frame, where columns are samples and rows are markers.
driftcg	List of drifting CpGs for which to calculate residuals.
chrage	Sample chronological (patient) age, in same order as columns in df.
showStatusDrift	Whether to show detailed output of calculation status.
keepres	Whether to retain CpG-level residuals used in calculation of mean.

Value

Returns age estimate and specified additional information.

Author(s)

Sean K Maden

Examples

## The function is currently defined as
function(df, driftcg,chrage,showStatusDrift=TRUE,keepres=FALSE){
  # note: chrage sample age order must correspond to order of columns in df
  message("Beginning Drift CpG residuals calculations...")

  dfdrift <- df[rownames(df) 
  reslist <- c()
  for(i in 1:nrow(dfdrift)){
    reslist[[i]] <- lm(dfdrift[i,]~chrage)$residuals
    names(reslist)[i] <- rownames(dfdrift)[i]
    if(showStatusDrift){
#      message("Drift Residuals Status: finished ",i," or ",round((100*(i/nrow(dfdrift))),3),"%.")
    }
  }

  xvecti <- c(rep(0,ncol(dfdrift)))
  for(i in 1:ncol(dfdrift)){
    xvecti[i] <- mean(sapply(reslist, "[", i))
    names(xvecti)[i] <- colnames(dfdrift)[i]
  }
  driftlm <- data.frame(Mean.Residual=xvecti,Chr.Age=chrage)
  rownames(driftlm) <- colnames(dfdrift)

  return.list <- list(driftlm)

  if(keepres){
    return.list <- append(return.list,list(reslist))
    names(return.list)[[length(return.list)]] <- "All.Drift.Residuals"
  }
  return(return.list)
}

metamaden/cgageR documentation built on May 20, 2019, 5:07 p.m.