Hi,

recently I came across this tool and wanted to give it a try. When I used a bed file it gave the following error

Error in quantile.default(bb) :
  missing values and NaN's not allowed if 'na.rm' is FALSE
In addition: Warning messages:
1: In CREAM(in_path = "my_70percent_overlap.bed", out_path = "./my_70percent_overlap_CORES_CREAM.bed",  :
  There are chromosome with low number of regions and you may get error. The reason is that highest Order may become larger than the number of regions in a chromosome. Hence, there will not be enough regions in that chromosome for clustering.
2: In log(SortedWindow_Vec[-RemovePeaks]) : NaNs produced

How would you suggest to deal with it?

Thank you.

DESCRIPTION file requires the following info:

• Title: (acronym has to be described). Make sure to type out the title in Title Case
• Description: (detailed description of what the package does)
• Imports: packages that provide functions, methods, or classes that are used inside your package name space.
• Suggests: packages used in vignettes or examples, or in conditional code.
• biocViews: “keywords” used to describe a given package (see https://www.bioconductor.org/developers/how-to/biocViews/)

See https://www.bioconductor.org/developers/package-guidelines/#dependencies for more on R package guidelines.

Template for unit tests has already been provided -> https://github.com/bhklab/CREAM/tree/master/tests/testthat

I have been trying to implement LOCKs using this repository and your codeocean link from your recent paper (https://codeocean.com/capsule/6911149/tree/v1).

I understand that the CREAM_Cutoff is the iterative optimization of WScutoff. The new CREAM parameter optimize=TRUE evokes ElementRecog_freecutoff instead of ElementRecog. However, there is also a WindowSizeRecog_freecutoff, but I do not see any scripts evoking it. Was this not implemented in the final version of the CREAM update?

Template for vignettes has already been provided -> https://github.com/bhklab/CREAM/tree/master/vignettes

A general theory question: can/should the updated CREAM framework (used to define LOCKs) also be used to define COREs on accessibility data?

From my understanding, the CREAM_Cutoff iterations just identify an optimal WScutoff threshold, i.e. the optimal window size within the empirical distribution. This suggests to me that this optimization would also improve detecting canonical COREs as well. Further this framework could be practically applied to cluster peaks from almost any chromatin feature (not just modified lysine residues), I would guess.

Existing documentation is great, but we still need to:

• Describe what each function returns
• Add an example of how each function is used

The object 'Outliers' is called in WindowSizeRecog_freecutoff but not defined. CREAM() with current build errors because 'Outliers' is not defined.

In WindowSizeRecog, 'Outliers' is defined as: Outliers <- which(bb < TightReg). May need to just add this line to _freecutoff version?

I tried to use CREAM with output from Genrich (narrowPeaks) format. I run into this error and unsure how to proceed. Would you mind helping me?
#CREAMfunction
IdentifiedCOREs <- CREAM(in_path=system.file("extdata", "816_775.narrowPeaks.bed", package = "CREAM"), MinLength = 1000, peakNumMin = 2 )
Error in read.table(in_path, sep = "\t") : no lines available in input
In addition: Warning message:
In file(file, "rt") :
file("") only supports open = "w+" and open = "w+b": using the former

The command
write.table(CombinedData , file = out_path,
row.names=FALSE, col.names = FALSE, quote = FALSE, sep = "\t")
at the end of CREAM function should be replaced by
return(CombinedData)

The example and description of the function should be modified accordingly.