mtmorgan
commented
6 years ago 1) define a named list, maybe in a file R/errors.R, e.g.,
Errors <- list(
## ExCluster
missing_exon_count_matrix =
"You did not provide a normalized exon count matrix to the
exonCounts argument. Please obtain normalized exon counts
from processCounts() and assign the resutls to exonCounts.
For example, if you followed the vignette, you should have a
normCounts variable. In this case, you could specify:
exonCounts=normCounts",
exon_counts_column_mismatch =
"Your number of exonCounts columns did not equal the length of
your number of condition identifiers. Please ensure that you
entered correct condition identifiers to cond.Nums, such as
cond.Nums=c(1,1,1,2,2,2). Additionally, please also ensure
your normalized counts provided to exonCounts have only count
data columns. All Gene:ExonBin identifiers should be provided
as rownames. If you are unsure, please refer to the vignette."
)use the list in your R code, e.g.,
if (is.null(exonCounts) == TRUE){
stop(Errors$missing_exon_count_matrix)
}
### ensure the number of columns in the exonCounts parameter equal the number of cond.Nums
if (ncol(exonCounts) != length(cond.Nums)){
stop(Errors$exon_counts_column_mismatch)
}Fancier solutions are possible.
2) I'm not exactly sure what you mean, but starting from the top of ExCluster.R() I'd create a file R/utilities.R with things like
#### Geometric mean function
gm_mean = function(x, na.rm=TRUE){
exp(sum(log(x[x > 0]), na.rm=na.rm) / length(x))
}
#Compute effect size
Calc.effect.size <- function(Mean1, Mean2, Var1, Var2){
(Mean1 - Mean2) / (sqrt((Var1 + Var2) / 2))
}and simply remove the definitions from ExCluster -- when you use gm_mean() in the body of ExCluster, R will look for a gm_mean() in the body of the function. It won't find it, so it will for a function gm_mean() in the environment where the ExCluster() function was defined. This environment is (inside) the package name space, where it will find gm_mean() you defined in utilities.R.
The third function in ExCluster is more problematic
Determine_HC_Distance <- function(x){
which(HC_cutree%in%x)
}because it references HC_cutree, which is not an argument. The correct solution is to update the function so that all symbols in the body of the function are either arguments or symbols from your package or the symbols your package imports. So
Determine_HC_Distance <- function(HC_cutree, x) {
which(HC_cutree %in% x)Of course the code where you use this function
HC_indices <- lapply(seq(NumClusters),FUN=Determine_HC_Distance)has to be modified
HC_indices <- lapply(seq(NumClusters),FUN=Determine_HC_Distance, HC_cutree = HC_cutree)The diversity of naming conventions in these functions is also confusing to an outsider -- is there some kind of code, where functions with _ do one thing, with . do another, etc? But probably it is just inconsistency and the code would really benefit from more consistency.
I see a lot of loops in your code, like
Generate_ES_dists <- function(Mean1,Var1){
x=length(Mean1)
ES_mat <- matrix(0,nrow=x,ncol=x)
# fill in nth row and jth column data for ES difference matrix
for (n in seq(x-1)){
for (j in (n+1):x){
ES_mat[n,j] <- Calc.effect.size(Mean1[n],Mean1[j],Var1[n],Var1[j])
ES_mat[j,n] <- ES_mat[n,j] * -1
}
}
return(ES_mat)
}If you think about it, Calc.effect.size() works as well on scalar values as on vectors of values. Also, a matrix can be subset by a two-column matrix to update selected rows and columns. So one could first generate the vectors n and j, and then avoid most of the iteration
Generate_ES_dists_1 <- function(Mean1, Var1) {
x <- length(Mean1)
ES_mat <- matrix(0, nrow=x, ncol=x)
n <- rep(seq_len(x - 1), rev(seq_len(x - 1)))
j <- unlist(lapply(seq_len(x - 1) + 1, seq, to = 10))
idx <- cbind(n, j)
ES_mat[idx] <- Calc.effect.size(Mean1[n],Mean1[j],Var1[n],Var1[j])
ES_mat[ idx[, c(2, 1)] ] <- ES_mat[idx] * -1
ES_mat
}This produces the same result but is about 10x faster than the original; this is more-or-less the usual case with vectorized code versus iteration.
RMTbioinfo
lshep
lawremi
bioc-issue-bot
Update the following URL to point to the GitHub repository of the package you wish to submit to Bioconductor
Confirm the following by editing each check box to '[x]'
[x] I understand that by submitting my package to Bioconductor, the package source and all review commentary are visible to the general public.
[x] I have read the Bioconductor Package Submission instructions. My package is consistent with the Bioconductor Package Guidelines.
[x] I understand that a minimum requirement for package acceptance is to pass R CMD check and R CMD BiocCheck with no ERROR or WARNINGS. Passing these checks does not result in automatic acceptance. The package will then undergo a formal review and recommendations for acceptance regarding other Bioconductor standards will be addressed.
[x] My package addresses statistical or bioinformatic issues related to the analysis and comprehension of high throughput genomic data.
[x] I am committed to the long-term maintenance of my package. This includes monitoring the support site for issues that users may have, subscribing to the bioc-devel mailing list to stay aware of developments in the Bioconductor community, responding promptly to requests for updates from the Core team in response to changes in R or underlying software.
I am familiar with the essential aspects of Bioconductor software management, including:
For help with submitting your package, please subscribe and post questions to the bioc-devel mailing list.