rsa: Perform RSA ranking on the screening results.
In cellHTS2: Analysis of cell-based screens - revised version of cellHTS

Description Usage Arguments Details Value Author(s) References See Also Examples

The RSA method ranks the resulting hit list of a screening experiment, taking into account the design of the screening library (i.e., multiple probes targeting the same effector molecule).

1	rsa(x, geneColumn="GeneID", lowerBound=0, upperBound=1, reverse=FALSE, drop=FALSE)

`x`	Object derived from class `cellHTS`.
`geneColumn`	The name of the well annotation column to be used for the grouping of effector molecules and probes.
`lowerBound`	The lower boundary parameter for the RSA algorithm.
`upperBound`	The upper boundary parameter for the RSA algorithm.
`reverse`	Boolean. Reverse the ranking.
`drop`	Boolean. Drop all probes from the analysis for which no effector molecule is defined.

The input argument x has to be a cellHTS2 object which has been scored, summarized and annotated. For details on the RSA algorithm please see the publication referenced below.

A data.frame with the following columns:

`Value of argument geneColumn:`	the target molecule identifier.
`Plate:`	the plate identifier.
`Well:`	the well identifier.
`Score:`	the probe score in the screen.
`RSARank:`	the computed RSA rank.
`ScoreRank:`	the rank based on a simple cutoff scheme.
`PValue:`	the computed RSA $p$-value.
`RSAHit:`	the RSA hit flag.
`#HitWell:`	the number of probes counted as positive RSA hits for a given target molecule.
`#TotalWell:`	the total number of probes for a given target molecule.
`%HitWell:`	the percentage of postive hits for a given molecule.

Florian Hahne florian.hahne@novartis.com

Renate Koenig, Chih-yuan Chiang, Buu P Tu, S Frank Yan, Paul D DeJesus, Angelica Romero, Tobias Bergauer, Anthony Orth, Ute Krueger, Yingyao Zhou & Sumit K Chanda: A probability-based approach for the analysis of large-scale RNAi screens

NATURE METHODS | VOL.4 NO.10 | OCTOBER 2007 | 847

cellHTS

data(KcViabSmall)
KcViabSmall <- scoreReplicates(KcViabSmall, sign="-", method="zscore")
KcViabSmall <- summarizeReplicates(KcViabSmall, summary="mean")
ranking <- rsa(KcViabSmall)
head(ranking)