reportHTML.spikeTrain: Performs Basic Spike Train Analysis and Generates a Report in...
In STAR: Spike Train Analysis with R

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

Performs a "standard" analysis on a spikeTrain object, computing some cross-correlation statistics if additional spikeTrain objects are provided, writes results to disk and generates a report in html format.

## S3 method for class 'spikeTrain'
reportHTML(object, filename, extension = "html",
           directory = getwd(), Title, forceTT = TRUE,
           digits = 3, timeUnit = "s", otherST,
           laglim = c(-0.1, 0.1),
           cch = c("both", "scch", "cch"),
           method = c("gsslockedTrain0", "gsslockedTrain", "gamlockedTrain"),
           doGamCheck = FALSE, k = 100, bs = "tp",
           nbEvtPerBin = 10, ...)

`object`	a `spikeTrain` object.
`filename`	a character string. The generic name of all the files (html, png as well as `R` data files which will be generated. See also `HTMLInitFile`.
`extension`	see `HTMLInitFile`.
`directory`	the full or relative path to the directory where the results are going to be stored. See also `HTMLInitFile`.
`Title`	See `HTMLInitFile`. If missing a default value baed on `filename` is provided.
`forceTT`	Should a time transformation be performed and the `compModels` plots be generated even if none of the six renewal models fits the data?
`timeUnit, digits`	see `summary.spikeTrain`.
`otherST`	a named list of `spikeTrain` objects from simultaneously recorded neurons or nothing.
`laglim`	see `lockedTrain`.
`cch`	if `otherST` is given (ie, not missing) cross-intensity plots will be made using the neuron of `spikeTrain` as a reference. Should smooth version of the cross-intensity be computed (`"scch"`), a "classical" one (`"cch"`) or both (`"both"`). Only the first element of `cch` is used.
`method`	A character string, the name of the function used to generate the smooth cross-correlation histograms, one of: `gsslockedTrain`, `gsslockedTrain0`, `gamlockedTrain`.
`doGamCheck`	if smooth estimates are requested and `method` is set to `gamlockedTrain`, should function `gam.check` be used on them?
`k`	see `gamlockedTrain`.
`bs`	see `gamlockedTrain`.
`nbEvtPerBin`	a number of event per bin used in a way similar to the argument with the same name in `jpsth` when a bining is used for pre-processing.
`...`	Passed to `gsslockedTrain`, `gsslockedTrain0`, `gamlockedTrain`.

A spike train plot (plot.spikeTrain) is performed first. The summary (summary.spikeTrain) is computed next and part of its output is written to the html file. The renewal tests are then carried out and their results added (renewalTestPlot). The six duration distributions are fitted (compModels with argument plot set to FALSE) and the best one is used to apply a time transformation to spikeTrain. The Ogata's tests are applied (summary.transformedTrain) and if they are all within the 99% confidence interval, the result of the transformation is plotted (plot.transformedTrain) as well as all the Q-Q plots of compModels. If forceTT is set to TRUE (default), then these last two plots are added even if the best model does not pass the tests.

If other spikeTrain objects are provided as a named list via argument otherST, then cross-correlation/cross-intensity functions are estimated; Two estimations methods are available, the classical histogram and a smooth version of it. Argument cch controls if a single estimation is performed or if both are performed. If the smooth version is requested a summary of the gssanova, gssanova0 or gam fit is printed (depending on the chosen value for argument method). Moreover if argument doGamCheck is set to TRUE (and if method is set to gamlockedTrain) then check plots (gam.check) are added to the report.

A R data file (filename.rda) is also generated with the following objects:

cm: the result of compModels.
bestFit: the durationDistribution object returned obtained by fitting the best model among the 6.
Lambda: the integrated intensity of spikeTrain with the best model.
fct: the matched call.
cchL: if other trains were provided and if argument cch was set to "both" or to "cch". A list with as many components as the otherST argument. Each component is the a hist.lockedTrain object.
scchL: if other trains were provided and if argument cch was set to "both" or to "scch". A list with as many components as the otherST argument. Each component is the a gsslockedTrain, gsslockedTrain0 or gamlockedTrain object.

Nothing is returned, an html file and figures in png format are written to disk together with the R variables generated during the analysis.

Christophe Pouzat christophe.pouzat@gmail.com

as.spikeTrain, plot.spikeTrain, summary.spikeTrain, renewalTestPlot, plot.spikeTrain, compModels, transformedTrain, plot.transformedTrain, summary.transformedTrain, gssanova, gssanova0, gam, gam.check, lockedTrain, gsslockedTrain, gsslockedTrain0, gamlockedTrain

## Not run: 
## load e070528spont data set
data(e070528spont)
## perform a standard analysis on neuron 1, looking for cross-correlations
## with the 3 other neurons up to lag +/- 250 ms.
## Store the results under the generic name: e070528spontN1
reportHTML(e070528spont[["neuron 1"]],"e070528spontN1",otherST=e070528spont[-1],laglim=c(-1,1)*0.25,forceTT=FALSE)
## Neuron 1 of e070528spont is exceptional in that it can be well
## described by a renewal process...
## End(Not run)