Function to compute sliding (regularized) one- or two-sample T statistics on a tiling expression set.

1 |

`xSet` |
Object of class |

`probeAnno` |
Environment holding the genomic positions of probes in the ExpressionSet |

`allChr` |
Character vector of all chromosomes in genome |

`test` |
character; one of |

`grouping` |
factor vector of length equal to number of samples,
not required if |

`winHalfSize` |
Half the size of the window centered at a probe position, in which all other probes contribute to the calculation of the mean and standard deviation. |

`min.probes` |
integer; if less probes are in the sliding window, NA instead of the mean and sd is returned. This is meant to avoid to computing non-meaningful means and standard deviations. If unwanted, set this to 1 or less |

`checkUnique` |
logical; indicates whether the uniqueness indicator of probe matches from the probeAnno environment should be used. |

`uniqueCodes` |
numeric; which numeric codes in the chromosome-wise match-uniqueness elements of the probeAnno environment indicate uniqueness? |

`verbose` |
logical; detailed progress output to STDOUT? |

An object of class `ExpressionSet`

, holding the T statistics
values for the probes of the supplied ExpressionSet. The number of
results samples is the number of levels in the supplied factor
`grouping`

.

Joern Toedling

`sliding.meansd`

1 2 3 4 5 6 7 8 9 10 11 12 | ```
exDir <- system.file("exData",package="Ringo")
load(file.path(exDir,"exampleProbeAnno.rda"))
load(file.path(exDir,"exampleX.rda"))
tX <- computeSlidingT(exampleX, probeAnno=exProbeAnno,
allChr=c("9"), winHalfSize=400)
sampleNames(tX) <- "t-Stat_Suz12vsTotal"
if (interactive()){
grid.newpage()
plot(cbind2(exampleX, tX), exProbeAnno, chrom="9",
xlim=c(34318000,34321000), ylim=c(-2,8.5), gff=exGFF,
paletteName="Paired")
}
``` |

Questions? Problems? Suggestions? Tweet to @rdrrHQ or email at ian@mutexlabs.com.

All documentation is copyright its authors; we didn't write any of that.