SummarizedExperiment::cbind

[tiagochst]

Hello,

this is just a suggestion. If a have to SE, with the same data, but one has a metadata column that the other does not have.

The package gives the following error:

Error in map_x_colnames_to_object_colnames(colnames(object)) : 
  the DataFrame objects to rbind must have the same number of columns

But instead of using a rbind wouldn't it be better to use a rbind.fill (https://www.rdocumentation.org/packages/plyr/versions/1.8.4/topics/rbind.fill)
in this part code

SummarizedExperiment/R/SummarizedExperiment-class.R

Line 711 in 01aa07a

colData <- do.call(rbind, lapply(args, colData))

In that way, it will populate the metadata with NA and shows show a warning instead.
I am not sure if it was discussed before. Otherwise, it would be nice to specify more clearly that the problem is within the metadata.

Here is an example of the problem if we add year_to_death column.

nrows <- 200; ncols <- 6
counts <- matrix(runif(nrows * ncols, 1, 1e4), nrows)
rowRanges <- GRanges(rep(c("chr1", "chr2"), c(50, 150)),
                     IRanges(floor(runif(200, 1e5, 1e6)), width=100),
                     strand=sample(c("+", "-"), 200, TRUE),
                     feature_id=sprintf("ID%03d", 1:200))
colData <- DataFrame(Treatment=rep(c("ChIP", "Input"), 3),
                     row.names=LETTERS[1:6])
rse <- SummarizedExperiment(assays=SimpleList(counts=counts),
                            rowRanges=rowRanges, colData=colData)

colData <- DataFrame(Treatment=rep(c("ChIP", "Input"), 3),
                     years_to_death = 1:6,
                     row.names=LETTERS[1:6])
rse2 <- SummarizedExperiment(assays=SimpleList(counts=counts),
                            rowRanges=rowRanges, colData=colData)

SummarizedExperiment::cbind(rse2, rse)

[wjyzidane]

I got the same error. Is there a way to remove a specific column in ColData?

[lawremi]

Sure, just replace the column with NULL like you would do with an R data.frame:

colData(se)$foo <- NULL

In my opinion, the current behavior is fine. If there are inconsistencies in the metadata, the user should do something explicit. For consistency, safety, etc, cbind() should just do rbind() on the column metadata. Maybe there could be a merge() method that is more forgiving.

[wjyzidane]

Thanks very much!

…

On Aug 1, 2019, at 10:59 PM, Michael Lawrence ***@***.***> wrote: Sure, just replace the column with NULL like you would do with an R data.frame: colData(se)$foo <- NULL In my opinion, the current behavior is fine. If there are inconsistencies in the metadata, the user should do something explicit. For consistency, safety, etc, cbind() should just do rbind() on the column metadata. Maybe there could be a merge() method that is more forgiving. — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or mute the thread.