GETPPLR(1)

πŸ“¦
version 0.7.5 Β· 4 versions
, in ubuntu jammy, ubuntu noble

🌸 getPPLR(1) β€” the manual page, but make it a group chat 🌸

okay bestie, welcome to getPPLR, from section 1 of the manual (that's User Commands, aka "stuff you just type, no root required πŸ’…"). This is genuinely one of those tools that sounds scary and is actually just doing math on your behalf while you sip your lil drink. Let's get into it.


✨ NAME ✨

getPPLR β€” compute PPLR from MCMC expression samples πŸ’–

That's it, that's the whole job. PPLR. From MCMC. From your expression samples. Iconic behavior, honestly.


πŸ¦‹ SYNOPSIS πŸ¦‹

getPPLR -o <outFileName>  [OPTIONS] [sampleFile-C1] [sampleFile-C1]

so yeah, you need -o <outFileName> (she's required, she's the main character πŸ‘‘) and then whatever else you want after that. Note that the sample file placeholders are literally written as [sampleFile-C1] [sampleFile-C1] β€” yes, twice, same name, that's what the page says, and I'm not about to start editing the sacred texts. It's giving "two conditions, put your files in order" vibes. 🫢


🌈 DESCRIPTION 🌈

Computes PPLR from MCMC expression samples. 🧬✨

(the probability of second condition being up-regulated)

and it also computes log2 fold change with confidence intervals, plus the condition mean log expression. So like, three little treats in one command. Obsessed.

The [sampleFiles] should contain transposed MCMC samples from different conditions. Transposed, bestie. Do not come in here with untransposed samples, it's not the vibe. πŸ™ˆ


πŸ’œ OPTIONS πŸ’œ

  • --help β€” show this help information. The "hello, what does anything do" button. You love to see it. πŸŽ€
  • -d, --distribution β€” produce the whole distribution of differences. Not just the summary, the entire spread. default: Off ✨
  • --inputIsLogged β€” tells the tool your input expression estimates are on a log scale. Also: you do not need to use this with data generated by BitSeq-0.5.0 and above, so if that's you, chill. (default: Off) 🌸
  • -o <outFileName>, --outFile=<outFileName> β€” name of the output file. This one's mandatory, no skipping, she's the whole point. πŸ’…
  • -s <selectFileName>, --selectFile=<selectFileName> β€” a file containing a list of selected transcript IDs (zero based), and only those get reported. Heads up though: this only works with the --distribution option. So it's a package deal, don't try to run it solo. πŸ¦‹
  • --subSample=<subSample> β€” sub-sample the distributions using a given fraction of expression samples. (default: 1) πŸ“
  • -v, --verbose β€” verbose output. She talks more. (default: Off) πŸ”₯

πŸ‘‘ a lil sign-off πŸ‘‘

This whole thing is from getPPLR 0.7.5+dfsg, dated September 2021 β€” so if you're out here in the future wondering why something looks a lil retro, that's why, and she's still serving πŸ’– no cap. Now go run your samples, bestie, and remember: transposed, -o always, -s only with -d. periodt. 🌟

πŸ“¦ where this page comes from