confIntGPS-methods: Method for calculating GPS values and their 95% bootstrap...
In Rtreemix: Rtreemix: Mutagenetic trees mixture models.

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

The method first calculates the genetic progression score (GPS) for the patterns in a given dataset data based on a fitted mutagenetic trees mixture model with K components. The data and K have to be specified. Then, it derives a 95% confidence intervals for the GPS values with bootstrap analysis.

1	confIntGPS(data, K, ...)

`data`	An `RtreemixData` object containing the samples (patterns of genetic events) for which the GPS values and their bootstrap confidence intervals are to be calculated. The number of genetic events should NOT be greater than 20.
`K`	An `integer` larger than 0 specifying the number of branchings in the mixture model.
`...`	`sampling.mode` is a `character` that specifies the sampling mode ("constant" or "exponential") used in the waiting time simulations. Its default value is "exponential". `sampling.param` is a `numeric` that specifies the sampling parameter corresponding to the sampling mode given by `sampling.mode`. Its default value is 1. `no.sim` is an `integer` larger than 0 giving the number of iterations for the waiting time simulation. Its default values is 10000. `B` is an `integer` larger than 0 specifying the number of bootstrap samples used in the bootstrap analysis. Its default value is 1000. `equal.star` is a `logical` specifying whether to use equal edge weights in the noise component. The default value is `TRUE`. When you have few data samples always use its default value (`TRUE`) to ensure nonzero probabilities for all possible patterns (sets of events).

The function returns an object from the RtreemixGPS class that containes the calculated GPS values, their 95% confidence intervals, the model used for the computation, the data, and so on (see RtreemixGPS-class). The GPS values are represented as a numeric vector with length equal to the number of samples in data. Their corresponding confidence intervals are given in a matrix with two columns.

The data for which the GPS values and their corresponding confidence intervals are to be calculated should not have more than 20 genetic events. The reason for this is that the number of all possible patterns for which the GPS values are calculated during a computationally intensive simulations is in this case $2^20$. This demands too much memory. The GPS examples are time consuming. They are commented out because of the time restrictions of the check of the package. For trying out the code please copy it and uncomment it.

Jasmina Bogojeska

RtreemixGPS-class, gps-methods, RtreemixData-class, RtreemixModel-class, fit-methods

## Create an RtreemixData object from a randomly generated RtreemixModel object.
#rand.mod <- generate(K = 2, no.events = 7, noise.tree = TRUE, prob = c(0.2, 0.8))
#data <- sim(model = rand.mod, no.draws = 400)

## Create an RtreemixGPS object by calculating GPS values for a given dataset
## and their 95% confidence intervals using the bootstrap method.
#modGPS2 <- confIntGPS(data = data, K = 2, B = 100) ## time consuming computation
#show(modGPS2)

## See the GPS values for the object modGPS2 and their confidence intervals.
#GPS(modGPS2)
#gpsCI(modGPS2)

## See data.
#getData(modGPS2)