cellHTS (version 1.42.0)

scoreReplicates: Scores normalized replicate values given in a cellHTS object

Description

This function scores the normalized replicate values stored in slot assayData of a cellHTS object. Current available options are to take the z-score value or the per-replicate normalized percent inhibition (NPI). Data are stored in slot assayData overridding its current content.

Usage

scoreReplicates(object, sign="+", method="zscore", ...)

Arguments

object

an object of class cellHTS that has already been normalized.

sign

a character string, either "+" (default) or "-", which corresponds to multiplying the data by +1 or -1, respectively, after applying the scoring method specified by argument method. See details.

method

a character string indicating which method to use to score the replicate measurements. Available options are "none", "zscore" (default), "NPI". See details.

...

additional parameters required by some of the methods chosen in method.

Value

• A cellHTS object with its slot assayData replaced with the scored values (same dimension).

  Important: Note that the processing state "scored" of the cellHTS object is only updated to TRUE after summarizing the replicates, which is the next preprocessing step (see summarizeReplicates).

Details

This function scores the normalized values given in the slot assayData of object. Current availabe scoring methods are:

• method="none", no scoring is applied.
• method="zscore"(robustz-scores), for each replicate, this is calculated by subtracting the overall median from each measurement and dividing the result by the overallmad. These are estimated for each replicate by considering the distribution of intensities (over all plates) in the wells whose content is annotated assample.
• method="NPI"(normalized percent inhibition applied in a per-replicate basis, i.e. using the overall mean of positive and negative controls across all plates of a given replicate), for each replicate, this method consists of subtracting each measurement from the average of the intensities on the positive controls (taken across all plates), and this result is then divided by the difference between the averages of the measurements on the positive and the negative controls (taken across all plates). If this method is chosen, one may need to provide further arguments toscoreReplicates, namely, argumentsposControlsandnegControls. These arguments should be vectors of regular expression patterns specifying the name of the positive(s) and negative(s) controls, respectivey, as provided in the plate configuration file. The length of these vectors should match the current number of channels inobject(i.e.dim(Data(object))[3]). By default, ifposControlsornegControlsare not given,posandnegwill be taken as the name for the wells containing positive or negative controls. The content ofposControlsandnegControlsis passed toregexprfor pattern matching within the well annotation given in thefeatureDataslot ofobject(which can be accessed viawellAnno(object)) (see examples forsummarizeChannels).

After replicate scoring using the chosen method, the value given in sign ("+" or "-") is used to set the sign of the calculated scores. For example, with a sign="-", a strong decrease in the signal will be represented by a positive score, whereas setting sign="+", such a phenotype will be represented by a negative score. This option can be set to calculate the results to the commonly used convention.

References

Boutros, M., Bras, L.P. and Huber, W. (2006) Analysis of cell-based RNAi screens, Genome Biology 7, R66.

See Also

normalizePlates, summarizeChannels, summarizeReplicates.

Examples

data(KcViabSmall)
    x <- normalizePlates(KcViabSmall, scale="multiplicative", method="median", varianceAdjust="none")
    x <- scoreReplicates(x, sign="-", method="zscore")
    x <- summarizeReplicates(x, summary="min") # conservative approach