replist {qpcR} | R Documentation |
Starting from a 'modlist' containing qPCR models from single data, replist
amalgamates
the models according to the grouping structure as defined in group
. The result
is a 'modlist' with models obtained from fitting the replicates by pcrfit
.
replist(object, group = NULL, opt = FALSE, ...)
object |
an object of class 'modlist'. |
group |
a vector defining the replicates for each group. |
opt |
should model selection be applied to the final model? |
... |
other parameters to be supplied to mselect . |
As being defined by group
, the raw data of the curves are averaged and starting values for the averaged values are
calculated by optim
. Finally, a nls
model is built from the stack
ed raw data values using
these starting values. In contrast to curvemean
, the fluorescence values are averaged, not the cycle numbers.
Model selection can be applied to the final model by setting opt = TRUE
.
An object of class 'modlist' containing the replicate models of class 'nls'/'pcrfit'.
Andrej-Nikolai Spiess
ml <- modlist(reps, model = l4) rl <- replist(ml, group = gl(7, 4)) plot(rl) summary(rl[[1]]) ## optimize model based on Akaike weights rl2 <- replist(ml, group = gl(7, 4), opt = TRUE, crit = "weights")