confIntGPS-methods {Rtreemix} | R Documentation |
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.
## S4 method for signature 'RtreemixData, numeric': 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)