make p_bad_mappings - Githubissues

blahah / transrate

Understand your transcriptome assembly

http://hibberdlab.com/transrate

Other

100 stars 34 forks source link

make p_bad_mappings #178

Closed macmanes closed 8 years ago

macmanes commented 8 years ago

Write p_bad_mappings to assemblies.csv, like you do for p_good_mappings

blahah commented 8 years ago

this is just 1 - p_good_mappings - is it worth adding a column for it?

macmanes commented 8 years ago

you have p_everythingelse but I suppose you are right.. On the other hand, you already have like 33 columns, so what is one more...

blahah commented 8 years ago

I think for each thing we only report one way of expressing it, so e.g.

p_not_segmented but not p_segmented
p_bases_covered but not p_bases_uncovered
p_good but not p_bad

etc.

I would definitely prefer to have fewer columns :). However, it might be a good idea to explicitly state in the documentation that bad is always 1 - good, as there is no reason people should assume that is true.

blahah commented 8 years ago

we've stripped out some columns in v1.0.3, but we've still got too many. Closing this issue as we're not planning to put in p_bad_mappings, but we will be looking at reworking the output for clarity in the next major release.