R/hrdStats.R

In jpluta26/HRDX:

# ------------------------------------- hrd.stats ------------------------------------- #
# hrd.stats is a function to compute the three HRD metrics (HRD-LOH, HRD-NTAI, and
# HRD-LST), as well as total HRD and mean HRD. this simply wraps up the output in one
# dataframe- you can run all of these steps individually for error checking.

#
# input: seq.dat, (data.frame) with chromosome, start.pos, end.pos, CNt, alleleA, alleleB;
#         ploidy.dat (data.frame), the ploidy data
#         CN.dat (data.frame), copy number data
#         min.seg.size (integer), minimum segment size used in TAI calculations
#         scaleTotal (boolean), rescale HRD total to 0-100
# output: out, a data.frame with HRD metrics
hrd.stats <- function(seq.dat, ploidy.dat, CN.dat, min.seg.size = 6e06)
{
  
  seq.dat <- preprocessHRD(seq.dat)
  
  # raw data
  HRD.NTAIr <- getNTAI.raw( seq.dat, min.seg.size )
  HRD.TAIr  <- getTAI.raw(  seq.dat, min.seg.size )
  HRD.LSTr  <- getLST.raw(  seq.dat )
  HRD.LOHr   <- getLOH.raw(  seq.dat )
  HRD.LOHm  <- getLOH.norm( seq.dat, ploidy.dat)
  HRD.NTAIm <- getNTAI.norm( seq.dat, CN.dat, ploidy.dat, min.seg.size )
  HRD.TAIm  <- getTAI.norm(  seq.dat, CN.dat, ploidy.dat, min.seg.size )
  HRD.LSTm  <- getLST.norm( seq.dat, ploidy.dat )
  
  out = data.frame(HRD.LOHm   = HRD.LOHm, 
                   HRD.LOHr  = HRD.LOHr,
                   HRD.TAIr  = HRD.TAIr,
                   HRD.TAIRm = HRD.TAIm,
                   HRD.NTAIr = HRD.NTAIr,
                   HRD.NTAIm = HRD.NTAIm,
                   HRD.LSTr  = HRD.LSTr,
                   HRD.LSTm  = HRD.LSTm )
  
  # HRD.Score can be the total of any 3 metrics, raw or norm- this is the 'standard' score
  out$HRD.Score <- getHRD.Score( seq.dat, CN.dat, ploidy.dat, min.seg.size, scaleTotal = FALSE )
  
  return(out)
  
}
# ------------------------------------------------------------------------------- #