filterPolarity

[zeehio]

Thank you for your time and efforts on this package.

My apologies if I am overlooking something obvious (I am a newcomer here). If that is the case, feel free to tell me and I will close the issue.

I found out I needed a filterPolarity function and I wonder if it would be worth including a filterPolarity function in your MSnbase package. Here is my use case and a demo implementation:

Without polarity filtering

Loading a single sample and plotting a region, without filterPolarity, both polarities appear mixed:

# My apologies, I don't have permissions to submit data right now :-S
MSnbase::readMSData("1.mzXML") %>%
  MSnbase::filterRt(rt = c(500, 520)) %>%
  MSnbase::filterMz(mz = c(300, 400)) %>%
  MSnbase::plot(type = "XIC")

With polarity filtering

Based on MSnbase::filterMsLevel I wrote this filterPolarity function.

filterPolarity <- function(object, polarity.) {
  if (missing(polarity.)) return(object)
  polarity. <- as.numeric(polarity.)
  object[MSnbase::polarity(object) %in% polarity.]
}

Filtering the polarity I can easily see the positive and negative polarities:

MSnbase::readMSData("1.mzXML") %>%
  filterPolarity(polarity. = 1) %>%
  MSnbase::filterRt(rt = c(500, 520)) %>%
  MSnbase::filterMz(mz = c(300, 400)) %>%
  MSnbase::plot(type = "XIC")

MSnbase::readMSData("1.mzXML") %>%
  filterPolarity(polarity. = 0) %>%
  MSnbase::filterRt(rt = c(500, 520)) %>%
  MSnbase::filterMz(mz = c(300, 400)) %>%
  MSnbase::plot(type = "XIC")

Is there a reason why there is not a filterPolarity function in MSnbase? I feel I am missing something obvious, but I'd rather ask just in case...

Thanks

[lgatto]

Thank you for your report @zeehio. Firstly, I suppose we should indeed add a function filterPolarity.

I will let @jotsetung also chime in, put I spotted an issue in your function (which might be a false positive though):

filterPolarity <- function(object, polarity.) {
  if (missing(polarity.)) return(object)
  polarity. <- as.numeric(polarity.)
  object[MSnbase::polarity(dataset_not_dia) %in% polarity.] ## should be object, rather than dataset_no_dia
}

Could you confirm if that's the reason for the lack of effective filtering, or if it's only a typo.

[zeehio]

That dataset_not_dia should have been object, it was a copy-paste error on my side. Thanks for spotting it, I have updated the code above fixing it.

On my own code filterPolarity was working fine and the results match my expectations. I would contribute a pull request but I am just starting to get familiar with all those Bioconductor methods...

[lgatto]

I have added the method to MSnbase - only on github for now. And thank you for feature request.