| Title: | Optimal Row-Column Designs for Two-Colour cDNA Microarray Experiments |
|---|---|
| Description: | Computes A-, MV-, D- and E-optimal or near-optimal row-column designs for two-colour cDNA microarray experiments using the linear fixed effects and mixed effects models where the interest is in a comparison of all pairwise treatment contrasts. The algorithms used in this package are based on the array exchange and treatment exchange algorithms adopted from Debusho, Gemechu and Haines (2016, unpublished) algorithms after adjusting for the row-column designs setup. The package also provides an optional method of using the graphical user interface (GUI) R package tcltk to ensure that it is user friendly. |
| Authors: | Legesse Kassa Debusho, Dibaba Bayisa Gemechu, and Linda Haines |
| Maintainer: | Dibaba Bayisa Gemechu <[email protected]> |
| License: | GPL-2 |
| Version: | 1.0.0 |
| Built: | 2024-10-31 19:57:35 UTC |
| Source: | CRAN |
Computes the information matrix (C-matrix) for treatment effects under either the linear fixed effects model or the linear mixed effects model setting for a given row-column design.
cmatrcd.mae(trt.N, col.N, theta, des)cmatrcd.mae(trt.N, col.N, theta, des)
trt.N |
integer, specifying number of treatments, |
col.N |
integer, specifying number of arrays (columns), |
theta |
numeric, representing a function of the ratio of random array variance and random error variance. It takes any value between 0 and 1, inclusive. |
des |
matrix, a |
Returns a v x v treatment information matrix (C-matrix).
Legesse Kassa Debusho, Dibaba Bayisa Gemechu, and Linda Haines
Debusho, L. K., Gemechu, D. B., and Haines, L. M. (2016). Algorithmic construction of optimal block designs for two-colour cDNA microarray experiments using the linear mixed model. Under review.
Gemechu, D. B., Debusho, L. K., and Haines, L. M. (2015). A-and D-optional row-column designs for two-colour cDNA microarray experiments using linear mixed effects models. South African Statistical Journal, 49, 153-168.
optrcdmaeAT, fixparrcd.mae, intcrcd.mae
##Information matrix trt.N <- 3 col.N <- 3 theta <- 0.1 rcdes <- intcrcd.mae(trt.N = 3, col.N = 3) cmatrcd.mae(trt.N = 3, col.N = 3, theta = 0.1, des = rcdes)##Information matrix trt.N <- 3 col.N <- 3 theta <- 0.1 rcdes <- intcrcd.mae(trt.N = 3, col.N = 3) cmatrcd.mae(trt.N = 3, col.N = 3, theta = 0.1, des = rcdes)
Creates a GUI tcltk window that allow the users to set or fix values for the parametric combinations to compute optimal or near-optimal row-column designs.
fixparrcd.mae(Optcrit)fixparrcd.mae(Optcrit)
Optcrit |
character, specifying the optimality criteria to be used. |
fixparrcd.mae creates a pop-up GUI tcltk window that allow the users to set the parametric combinations to compute optimal or near-optimal row-column designs. The parameters include the number of treatments trt.N, number arrays col.N,
theta value theta, number of replications of the optimization procedure nrep and number of iterations required during exchange procedure itr.cvrgval.
Furthermore, on this window, the checkbox options that allow the users to choose whether or not they need to have the graphical
layout of the resultant optimal or near-optimal row-column design, to make a choice between the two-alternative algorithms
(treatment exchange and array exchange algorithms) and to print the summary of the resultant optimal or near-optimal row-column design on
R console directly are available.
After setting all the required parametric combinations and selecting the algorithm of interest,
clicking on the search button on the set parametric combinations tcltk window, similar to the results that
can be obtained when using the function optrcdmaeAT, the summary of the resultant optimal or near-optimal
row-column design will be saved in the current working directory in .csv format and it will also be displayed on R console with
graphical layout of the resultant optimal or near-optimal row-column designs.
The fixparrcd.mae function creates a pop-up tcltk window that allow the users to set
the parametric combinations to compute optimal or near-optimal row-column designs.
Legesse Kassa Debusho, Dibaba Bayisa Gemechu, and Linda Haines
optrcdmaeAT, mmenurcd.mae, tcltk, TkWidgets
Creates the graphical layout of resultant A-, MV-, D- or E-optimal or near-optimal row-column design on a separate pop-up GUI tcltk window.
graphoptrcd.mae(trt.N, col.N, theta, OptdesF, Optcrit, cbVal2)graphoptrcd.mae(trt.N, col.N, theta, OptdesF, Optcrit, cbVal2)
trt.N |
integer, specifying number of treatments, |
col.N |
integer, specifying number of arrays, |
theta |
numeric, representing a function of the ratio of random array variance and random error variance. It takes any value between 0 and 1, inclusive. |
OptdesF |
matrix, a |
Optcrit |
character, specifying the optimality criteria to be used. Thus, |
cbVal2 |
checkbox value. It takes a value of zero or one. The default value of |
Detail discussions concerning the constructions of a graphs can be found in igraph R package.
Returns the graphical layout of the resultant optimal or near-optimal row-column design 'OptdesF' on a separate pop-up window. Furthermore, the function graphoptrcd.mae saves the graphical layout of the resultant optimal or near-optimal row-column design in .pdf format in a working subdirectory.
When closing a pop-up window for graphical layout of the resultant designs (Graph plot), if the window is closed by
clicking on the red button with "X" sign (top-right), the warning message "Warning message: In rm(list = cmd, envir = .tkplot.env): object 'tkp ...' not found"
will occur in R console irrespective of what command is executed next. To resolve this warning message, click
on "close" menu that is located at the top-left of the graph plot pop-up window when closing this window.
Legesse Kassa Debusho, Dibaba Bayisa Gemechu, and Linda Haines
##To create the graphical layout of the D-optimal row-column design ##obtained using the treatment exchange algorithm for trt.N <- 10 #Number of treatments col.N <- 10 #Number of arrays theta <- 0.2 #theta value OptdesF <- rbind(1:10, c(2:10, 1)) #D-optimal design (loop design) graphoptrcd.mae(trt.N = 10, col.N = 10, theta = 0.2, OptdesF, Optcrit = "D", cbVal2 = 0)##To create the graphical layout of the D-optimal row-column design ##obtained using the treatment exchange algorithm for trt.N <- 10 #Number of treatments col.N <- 10 #Number of arrays theta <- 0.2 #theta value OptdesF <- rbind(1:10, c(2:10, 1)) #D-optimal design (loop design) graphoptrcd.mae(trt.N = 10, col.N = 10, theta = 0.2, OptdesF, Optcrit = "D", cbVal2 = 0)
Generates a random initial connected row-column design for a given number of arrays b of size k = 2 and the number of treatments v.
intcrcd.mae(trt.N, col.N)intcrcd.mae(trt.N, col.N)
trt.N |
integer, specifying number of treatments, |
col.N |
integer, specifying number of arrays, |
Returns a 2 x b connected row-column design with b arrays of size k = 2 and number of treatments v.
Legesse Kassa Debusho, Dibaba Bayisa Gemechu, and Linda Haines
Debusho, L. K., Gemechu, D. B., and Haines, L. M. (2016). Algorithmic construction of optimal block designs for two-colour cDNA microarray experiments using the linear mixed model. Under review.
Gemechu, D. B., Debusho, L. K., and Haines, L. M. (2014). A-optimal designs for two-colour cDNA microarray experiments using the linear mixed effects model. Peer-reviewed Proceedings of the Annual Conference of the South African Statistical Association for 2014 (SASA 2014), Rhodes University, Grahamstown, South Africa. pp 33-40, ISBN: 978-1-86822-659-7.
Gemechu, D. B., Debusho, L. K., and Haines, L. M. (2015). A-and D-optional row-column designs for two-colour cDNA microarray experiments using linear mixed effects models. South African Statistical Journal, 49, 153-168.
#Initial connected row-column design for trt.N <- 4 #Number of treatments col.N <- 4 #Number of arrays intcrcd.mae(trt.N = 4, col.N = 4)#Initial connected row-column design for trt.N <- 4 #Number of treatments col.N <- 4 #Number of arrays intcrcd.mae(trt.N = 4, col.N = 4)
Used to compute A-, MV-, D- or E-optimal or near-optimal row-column designs for two-colour cDNA microarray experiments under either the linear fixed effects model or the linear mixed effects model settings using either the array exchange or treatment exchange algorithms of Debusho, Gemechu and Haines (2016) after adjusting to the row-column setup.
optrcdmaeAT(trt.N, col.N, theta, nrep, itr.cvrgval, Optcrit = "", Alg = "", ...) ## Default S3 method: optrcdmaeAT(trt.N, col.N, theta, nrep, itr.cvrgval, Optcrit = "", Alg = "", ...) ## S3 method for class 'optrcdmaeAT' print(x, ...) ## S3 method for class 'optrcdmaeAT' summary(object, ...)optrcdmaeAT(trt.N, col.N, theta, nrep, itr.cvrgval, Optcrit = "", Alg = "", ...) ## Default S3 method: optrcdmaeAT(trt.N, col.N, theta, nrep, itr.cvrgval, Optcrit = "", Alg = "", ...) ## S3 method for class 'optrcdmaeAT' print(x, ...) ## S3 method for class 'optrcdmaeAT' summary(object, ...)
trt.N |
integer, specifying number of treatments, |
col.N |
integer, specifying number of arrays, |
theta |
numeric, representing a function of the ratio of random array variance and random error variance. It takes any value between 0 and 1, inclusive. |
nrep |
integer, specifying number of replications of the optimization procedure. |
itr.cvrgval |
integer, specifying number of iterations required for convergence during the exchange procedure. |
Optcrit |
character, specifying the optimality criteria to be used. |
x |
the object to be printed. |
object |
an object of class |
Alg |
character string used to specify the algorithm to be used. Possible values of |
... |
not used. |
optrcdmaeAT computes optimal or near-optimal row-column design for the two-colour cDNA microarray experiments
where the interest is in a comparison of all possible elementary treatment contrasts. The function computes A-, MV-, D- and E-optimal
or near optimal row-column designs via calling of eight sub-functions Aoptrcd.maeT, Aoptrcd.maeA,
MVoptrcd.maeT, MVoptrcd.maeA, Doptrcd.maeT, Doptrcd.maeA,
Eoptrcd.maeT and Eoptrcd.maeA, respectively. Each function requires an initial connected row-column designs,
generated using the function intcrcd.mae.
The minimum value of trt.N and col.N is 3 and trt.N should be less than or equal to col.N.
The linear fixed effects model results for given trt.N and col.N are obtained by setting theta = 0.0.
Alg specifies the array exchange and treatment exchange algorithm to be used that is adopted from Debusho, Gemechu and Haines (2016) after adjusting for the row-column designs setup. If Alg = "trtE", the function
optrcdmaeAT perform the treatment exchange procedure through deletion and addition of treatments at a time and selects a
design with best treatment exchange with respect to the optimality criterion value. If Alg = "arrayE", the function
optrcdmaeAT perform the array exchange procedure through deletion and addition of candidate arrays at a time and selects a
design with best array exchange with respect to the optimality criterion value.
nrep takes a value of greater than or equal to 2. However, to ensure optimality of the resultant design,
the nrep should be greater than or equal to 10 and in addition, as trt.N and col.N increase,
to ensure optimality of resultant design, it is advised to further increase the value of nrep
up to greater than or equal to 100. However, it has to be noted that as trt.N or col.N or
nrep or all of them increase, computer time required to generate optimal or near-optimal
row-column design increases.
itr.cvrgval number of iterations during exchange procedure. It takes a value between 2 and col.N. It is used
to speedup the computer search time by setting how long the user should wait for the exchange process to obtain any
different (if any) design than the one that was produced as the result of the preceding exchange of the current array in the initial
design with candidate array. This is mainly effective if col.N is very large. For example itr.cvrgval = 2, means the
exchange procedure will jump to the next array test if the exchange of the two preceding arrays with candidate arrays results with the
same efficient designs. The function will not give error message if the users set itr.cvrgval > col.N and it will automatically
set itr.cvrgval = col.N. The smaller the itr.cvrgval means the faster the exchange procedure is, but this will reduce the
chance of getting optimal row-column design and users are advised to set itr.cvrgval closer to col.N.
Remark: After the treatment exchange or array exchange procedure is completed, a dye-flip procedure is added to the internal functions of optrcdmaeAT stated above to further insure the optimality of the resulting optimal or near-optimal row-column designs. Thus, the procedure will flip (interchange) the treatments position within each array (column) and select the optimal dye-flip based on the optimality criteria of interest. This step is effective only for the large number of arrays and is efficient if itr.cvgval < col.N and there is a jump in the array exchange or treatment exchange procedure as stated above under the detail description of itr.cvrgval.
Returns the resultant A-, MV-, D- or E-optimal or near-optimal row-column design with its corresponding score value and parametric combination
saved in excel file in a working directory. In addition, the function optrcdmaeAT displays the graphical layout of the resultant
optimal or near-optimal row-column designs. Specifically:
call |
the method call. |
v |
number of treatments. |
b |
number of arrays. |
theta |
theta value. |
nrep |
number of replications of the optimization procedure. |
itr.cvrgval |
number of iterations required for convergence during the exchange procedure. |
Optcrit |
optimality criteria. |
Alg |
algorithm used. |
OptdesF |
a |
Optcrtsv |
score value of the optimality criteria |
file_loc, file_loc2
|
location where the summary of the resultant optimal or near-optimal row-column design is saved in .csv format. |
equireplicate |
logical value indicating whether the resultant optimal or near-optimal row-column design is equireplicate or not. |
vtrtrep |
vector of treatment replication of the resultant optimal or near-optimal row-column design. |
Cmat |
the C-matrix or treatment information matrix of the optimal or near-optimal row-column design. |
The graphical layout of the resultant optimal or near-optimal row-column design.
NB: The function optrcdmaeAT also saves the summary of the resultant optimal or near-optimal row-column design in .csv format in the working directory.
Furthermore, the function reports only one final optimal or near-optimal row-column design, however, there is a possibility
of more than one optimal or near-optimal row-column designs for a given parametric combination.
The function graphoptrcd.mae can be used to view and rearrange the graphical layout of the resultant
optimal or near-optimal row-column design on tcltk window. Alternative to the function optrcdmaeAT, a
GUI tcltk window can be used to generate optimal or near-optimal row-column designs, see mmenurcd.mae and fixparrcd.mae.
Legesse Kassa Debusho, Dibaba Bayisa Gemechu, and Linda Haines
Debusho, L. K., Gemechu, D. B., and Haines, L. M. (2016). Algorithmic construction of optimal row-column designs for two-colour cDNA microarray experiments using the linear mixed model. Under review.
Gemechu, D. B., Debusho, L. K., and Haines, L. M. (2014). A-optimal designs for two-colour cDNA microarray experiments using the linear mixed effects model. Peer-reviewed Proceedings of the Annual Conference of the South African Statistical Association for 2014 (SASA 2014), Rhodes University, Grahamstown, South Africa. pp 33-40, ISBN: 978-1-86822-659-7.
Gemechu, D. B., Debusho, L. K., and Haines, L. M. (2015). A-and D-optional row-column designs for two-colour cDNA microarray experiments using linear mixed effects models. South African Statistical Journal, 49, 153-168.
mmenurcd.mae, fixparrcd.mae, intcrcd.mae
##To obtain the A-optimal or near-optimal row-column design ##using treatment exchange algorithm, set trt.N <- 3 #Number of treatments col.N <- 3 #Number of arrays theta <- 0 #theta value nrep <- 5 #Number of replications itr.cvrgval <- 6 #Number of iterations required during the exchange procedure Optcrit <- "A" #Optimality criteria Alg <- "trtE" #Algorithm Aoptrcdes <- optrcdmaeAT(trt.N = 3, col.N = 3, theta = 0, nrep = 5, itr.cvrgval = 6, Optcrit = "A", Alg = "trtE") summary(Aoptrcdes)##To obtain the A-optimal or near-optimal row-column design ##using treatment exchange algorithm, set trt.N <- 3 #Number of treatments col.N <- 3 #Number of arrays theta <- 0 #theta value nrep <- 5 #Number of replications itr.cvrgval <- 6 #Number of iterations required during the exchange procedure Optcrit <- "A" #Optimality criteria Alg <- "trtE" #Algorithm Aoptrcdes <- optrcdmaeAT(trt.N = 3, col.N = 3, theta = 0, nrep = 5, itr.cvrgval = 6, Optcrit = "A", Alg = "trtE") summary(Aoptrcdes)
Functions for internal usage only.
## Computes A-optimal or near-optimal row-column designs ## using array exchange algorithm Aoptrcd.maeA(trt.N, col.N, theta, nrep, itr.cvrgval) ## Computes A-optimal or near-optimal row-column designs ## using treatment exchange algorithm Aoptrcd.maeT(trt.N, col.N, theta, nrep, itr.cvrgval) ## Computes MV-optimal or near-optimal row-column designs ## using array exchange algorithm MVoptrcd.maeA(trt.N, col.N, theta, nrep, itr.cvrgval) ## Computes MV-optimal or near-optimal row-column designs ## using treatment exchange algorithm MVoptrcd.maeT(trt.N, col.N, theta, nrep, itr.cvrgval) ## Computes D-optimal or near-optimal row-column designs ## using array exchange algorithm Doptrcd.maeA(trt.N, col.N, theta, nrep, itr.cvrgval) ## Computes D-optimal or near-optimal row-column designs ## using treatment exchange algorithm Doptrcd.maeT(trt.N, col.N, theta, nrep, itr.cvrgval) ## Computes E-optimal or near-optimal row-column designs ## using array exchange algorithm Eoptrcd.maeA(trt.N, col.N, theta, nrep, itr.cvrgval) ## Computes E-optimal or near-optimal row-column designs ## using treatment exchange algorithm Eoptrcd.maeT(trt.N, col.N, theta, nrep, itr.cvrgval)## Computes A-optimal or near-optimal row-column designs ## using array exchange algorithm Aoptrcd.maeA(trt.N, col.N, theta, nrep, itr.cvrgval) ## Computes A-optimal or near-optimal row-column designs ## using treatment exchange algorithm Aoptrcd.maeT(trt.N, col.N, theta, nrep, itr.cvrgval) ## Computes MV-optimal or near-optimal row-column designs ## using array exchange algorithm MVoptrcd.maeA(trt.N, col.N, theta, nrep, itr.cvrgval) ## Computes MV-optimal or near-optimal row-column designs ## using treatment exchange algorithm MVoptrcd.maeT(trt.N, col.N, theta, nrep, itr.cvrgval) ## Computes D-optimal or near-optimal row-column designs ## using array exchange algorithm Doptrcd.maeA(trt.N, col.N, theta, nrep, itr.cvrgval) ## Computes D-optimal or near-optimal row-column designs ## using treatment exchange algorithm Doptrcd.maeT(trt.N, col.N, theta, nrep, itr.cvrgval) ## Computes E-optimal or near-optimal row-column designs ## using array exchange algorithm Eoptrcd.maeA(trt.N, col.N, theta, nrep, itr.cvrgval) ## Computes E-optimal or near-optimal row-column designs ## using treatment exchange algorithm Eoptrcd.maeT(trt.N, col.N, theta, nrep, itr.cvrgval)
trt.N |
integer, specifying number of treatments, |
col.N |
integer, specifying number of arrays, |
theta |
numeric, representing a function of the ratio of random array variance and random error variance. It takes any value between 0 and 1, inclusive. |
nrep |
integer, specifying number of replications of the optimization procedure. |
itr.cvrgval |
integer, specifying number of iterations required for convergence during the exchange procedure. See |
These functions are handled via a generic function optrcdmaeAT. Please refer to the optrcdmaeAT documentation for details.
Legesse Kassa Debusho, Dibaba Bayisa Gemechu, and Linda Haines
Debusho, L. K., Gemechu, D. B., and Haines, L. M. (2016). Algorithmic construction of optimal block designs for two-colour cDNA microarray experiments using the linear mixed model. Under review.
Gemechu, D. B., Debusho, L. K., and Haines, L. M. (2014). A-optimal designs for two-colour cDNA microarray experiments using the linear mixed effects model. Peer-reviewed Proceedings of the Annual Conference of the South African Statistical Association for 2014 (SASA 2014), Rhodes University, Grahamstown, South Africa. pp 33-40, ISBN: 978-1-86822-659-7.
Gemechu, D. B., Debusho, L. K., and Haines, L. M. (2015). A-and D-optional row-column designs for two-colour cDNA microarray experiments using linear mixed effects models. South African Statistical Journal, 49, 153-168.