create_scExp(create_scDataset_raw()$mat, create_scDataset_raw()$annot) calls are broken

[hpages]

Hi,

Many man pages in the ChromSCape package contain the following call:

create_scExp(create_scDataset_raw()$mat,create_scDataset_raw()$annot)

As you can see the input to create_scExp() is obtained by 2 separate calls to create_scDataset_raw() which means that the mat and annot components are not coupled (create_scDataset_raw() is a non-deterministic function that returns a different result on each call). In particular the rownames on create_scDataset_raw()$mat are very unlikely to match those on create_scDataset_raw()$annot. Note that this matching is now enforced in SummarizedExperiment 1.23.2 (see).

This causes ChromSCape to fail on our daily builds for BioC devel. See https://bioconductor.org/checkResults/3.14/bioc-LATEST/ChromSCape/

The fix is to call create_scDataset_raw() only once:

raw <- create_scDataset_raw()
create_scExp(raw$mat, raw$annot)

Don't hesitate to ask on the bioc-devel mailing list if you have any question or need help with this.

Thanks, H.

[Pacomito]

Hello Hervé,

Thank you very much for the feedback and looking into this !

I didn't realize that create_scDataset_raw() with default parameters was actually non-deterministic.

I will change the call in the man pages using your fixed expression:

raw <- create_scDataset_raw()
create_scExp(raw$mat, raw$annot)

Thank you, Pacôme