STAT-VS-STAT(1)
π stat-vs-stat β the manual, but make it cute π
okay so hi bestie π«Ά welcome to stat-vs-stat.pl, which is genuinely one of those little scripts that just does the thing: it draws graphs of 1 stat vs another stat. that's it. that's the whole vibe. one stat. versus another stat. plot it. done. β¨
and yes this lives in the User Contributed Perl Documentation corner of the man pages, iykyk π
πΈ NAME πΈ
stat-vs-stat.pl β draws graphs of 1 stat vs another stat. periodt. π
β¨ SYNOPSIS β¨
cbtest.pl <input1> [input2 ...]
like. that's the whole invocation. give it one input, or a whole lil group chat of inputs. it's not picky. π
π¦ OPTIONS π¦
okay so first, the big picture before we get into the flags, because this part matters:
your input file should be some alignment that has a .rocr file alongside it β presumably generated by filter.pl --rocr π. and the output? it gets graphed to <basename>.cutoff.roc.png. so the name of your file, plus .cutoff.roc.png, and that's your picture. iconic.
now, the tea on sensitivity/specificity π: those need a false negative count, and that means they need a .filterstats file. no .filterstats, no sensitivity. that's just the rules bestie π
π Other options
--stat|predictor|xβ picks which stat you're using as the predictor. it's giving "which one goes on the x axis" energy.--samples|nβ sets the number of samples. the default is sample at each possible cutoff, so like, every single one. lowkey unhinged, highkey thorough. π--maxsamples|maxnβ puts a max onn/samples. a lil ceiling so it doesn't go feral. π--cleanβ forces a re-sampling of the.rocrfile. it's the "okay we're starting fresh" button. π§Ό--logβ can apply a log scale to the x or y or xy axes. your choice, bestie. β¨--fnβ can specify a FN count for calculating sensitivity. (fn = false negative, keep up π )--avgβ preload averages (normally found from filterstats). β οΈ and listen: averages are not drawn by default, so if you actually want them you have to specify--avgwith no args to enable them. no args!! okay!! π--noavgβ don't plot the line for averages. the anti---avg. simple as. π--lwdβ can specify line weight. thick line? thin line? main character line? your call. π--formatβ can specify the output file format. default ispng. πΌοΈ--pdfβ sets format topdf. a shortcut, and honestly? slay. π--nofstatsβ disables searching for filterstats statistics. like a lil "don't even look for them" π--nofnβ pretends not to have fn statistics, i.e. it skips sensitivity. the "we're ignoring that entire situation" flag. π--fstatsβ specify the.filterstatsfile to use. when you've got a specific one in mind. π«Ά
π reminder bestie: sensitivity/specificity = you need an FN count = you need a .filterstats file. don't come crying to me when it can't do the math π
πΈ sign-off πΈ
this lil page is living under perl v5.40.1, dated 2026-08-12, and it's part of the User Contributed Perl Documentation. that's the whole thing. go make some graphs, bestie, you're gonna ate πβ¨π
π¦ where this page comes from
- ubuntu stonking: murasaki-common 1.68.6-18 Β· GPL-3.0-or-later
- π girlified by deepseek-ai/DeepSeek-V4.1-Flash on : a language model's rewrite of the original, under its terms