rifi_preprocess: =========================================================================...

In CyanolabFreiburg/rifi: 'rifi' analyses data from rifampicin time series created by microarray or RNAseq

========================================================================= rifi_preprocess

rifi_preprocess wraps conveniently all pre-processing steps

Description

rifi_preprocess wraps the functions:

1. check_input

2. make_df

3. function_seg

4. finding_PDD

5. finding_TI

Usage

rifi_preprocess(
  inp,
  cores,
  FUN_filter = function(x) {     FALSE },
  bg = 0,
  rm_FLT = FALSE,
  thrsh_check = 0,
  dista = 300,
  run_PDD = FALSE,
  pen_PDD = 2,
  pen_out_PDD = 1,
  thrsh_PDD = 0.001,
  pen_TI = 10,
  thrsh_TI = 0.5,
  add = 1000
)

Arguments

inp	SummarizedExperiment: the input.
cores	integer: the number of assigned cores for the task.
FUN_filter	function: A function of x, returning a logical. x is the numeric vector of the intensity from all time points for a specific replicate.
bg	numeric: threshold over which the last time point has to be to be fitted with the above background mode.
rm_FLT	logical: remove IDs where all replicates are marked as filtered by the background check. Default is FALSE.
thrsh_check	numeric: the minimal allowed intensity for time point "0". Advised to be kept at 0! Default is 0.
dista	integer: the amount of nucleotides defining the gap. Default is 300.
run_PDD	logical: running the PDD flag function
pen_PDD	numeric: an internal parameter for the dynamic programming. Higher values result in fewer fragments. Advised to be kept at 2. Default is 2.
pen_out_PDD	numeric: an internal parameter for the dynamic programming. Higher values result in fewer possible outliers. Advised to be kept at 1. Default is 1.
thrsh_PDD	numeric: an internal parameter that allows fragments with slopes steeper than the threshold to be flagged with "PDD". Higher values result in fewer candidates . Advised to be kept at 0.001. Default is 0.001.
pen_TI	numeric: an internal parameter for the dynamic programming. Higher values result in fewer fragments. Advised to be kept at 10. Default is 10.
thrsh_TI	numeric: an internal parameter that allows fragments with a certain amount of IDs with higher relative intensities at time points later than "0" to be flagged as "TI". Higher values result in fewer candidates. -0.5 is 25 %, 0 is 50%, 0.5 is 75%. Advised to be kept at 0.5. Default is 0.5.
add	integer: range of nucleotides before a potential TI event where in IDs are fitted with the TI fit.

Details

rifi_preprocess allows for the optional integration of filter functions. Filter functions mark replicates with TRUE. Those are then not considered in the fit! FUN_filter is a general filter usually to exclude probes with low expression or "bad" patterns.

Value

The SummarizedExperiment object: checked, and with position, ID, intensity, probe_TI, position_segment, flag and filtration added to the rowRanges.

See Also

check_input

make_df

segment_pos

finding_PDD

finding_TI

Examples

data(example_input_minimal)
rifi_preprocess(
  inp = example_input_minimal, cores = 2, bg = 100, rm_FLT = FALSE,
  thrsh_check = 0, dista = 300, run_PDD = FALSE
  )
     

CyanolabFreiburg/rifi documentation built on May 7, 2023, 7:53 p.m.