2
$\begingroup$

As of 2019-01-28, there are 16 posts tagged with [filtering]. This is an ambiguous and, at least in some cases, unhelpful term. Some of these posts are about quality control, some are about ranking, some are about data sampling, and some are about eliminating redundancy.

I propose we eliminate the [filtering] tag. The [quality-control] tag can replace it as appropriate, while in many other cases it can probably just be removed.

$\endgroup$
2
$\begingroup$

I'd suggest the following changes

I'm not sure about the remaining posts. Some might be improved by dropping the [filtering] tag, or by replacing it with a different tag, or maybe it's the best term to use and we should simply clarify usage on the tag wiki.

$\endgroup$
  • $\begingroup$ I agree that filtering isn't a very good tag, but I don't think quality-control covers it either. For instance, removing specific reads from a fastq file doesn't seem to come under quality control (although the specific question you cite was indeed a quality issue). Maye we need a parsing tag? $\endgroup$ – terdon Feb 5 at 16:07
  • 1
    $\begingroup$ Well, it depends on why one is removing reads. If you want to reduce the size of the data, it's sampling. If you want to get rid of contaminant reads, it's quality control. If you want to more closely examine reads with particular nucleotide composition, that's...filtering? There are a lot of operations we can do on read data (or any data)--sorting/ranking, sampling, deduplicating, etc.--and not all of them need a tag. $\endgroup$ – Daniel Standage Feb 5 at 19:46
  • 2
    $\begingroup$ I guess? Maybe we should just stick to the relevant format tags (fastq, sam, vcf, gff etc) and leave it at that? $\endgroup$ – terdon Feb 5 at 19:47

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .