mcmc_analysis(1)

✨ mcmc_analysis — the manual page, but make it cute ✨

okay bestie, welcome to mcmc_analysis(1) 💖 this one lives in the General Commands Manual and it's literally the little analysis gremlin for your MCMC runs. let's get into it 🫶

---

🌸 NAME 🌸

mcmc_analysis — analysis of the MCMC. yeah. that's the whole name. she's simple, she's iconic, she does one thing 👑

---

💅 SYNOPSIS 💅

mcmc_analysis [OPTIONS] MCMC-data ...

---

🦋 DESCRIPTION 🦋

okay so. this program reads MCMC output — from one file, or from several files, whatever you've got — and those files come from programs in the BEEP package. and here's the thing bestie, the thing you cannot fumble: the same columns have to be present in every input file. no cap. if they don't match, that's on you 🙈

✨ Input format ✨

so the input is output from MCMC iterations, and it's formatted like this:

<likelihood> <tab> <iter> <tab> <params>

where, one by one:

• <likelihood> is the logarithm of the likelihood, in float format
• <tab> is tab whitespace separating the fields
• <iter> is an integer for the ordinal of the iteration
• <params> is a list of fields separated by semicolons, containing the parameters of the MCMC

and the MCMC params get their types and their names from the first line of the file, which looks like:

# L <tab> N <tab> [<name>(<type>)]+

the names ought to be unique, but honestly they don't have to be. lowkey chaotic of the format but okay. the <type> is one of these:

• float
• logfloat
• integer
• tree
• orthologypairs

and that type is what's used to infer how to parse and analyze the rest of the lines. so like, it's not decoration, it's the whole vibe-checker for the file 🧚

✨ Output format ✨

a report on the MCMC run, with posterior estimates of the parameters. slay.

---

🎀 OPTIONS 🎀

• -b [FLOAT|INT] — the burnin, bestie. you give either the percentage (0 <= x < 1) or the number of samples to discard (where x is an integer >= 1). default: 0.1.
• -p STRING — plot the parameter named <string>. the output is two columns: the iteration number, and that parameter's value in that iteration. plotting era 💖
• -i STRING — ignore the named parameter. go check the header line in your MCMC output for the column names, iykyk. you can name several columns at once with a comma-delimited format, like -i Length,Name. no spaces allowed between column names — not one, not ever, don't do it 🙈
• -t — output LaTex for the analysis. fancy academic queen behavior 👑
• -l — count and report the number of samples in the input file. just a lil headcount 🍓
• -mp NT — the maximum number of points to plot.
• -sp — explicitly indicate points in plots. be obvious about it, periodt.
• -coda — output file for the CODA package in R.
• -codatrees — same as -coda, but it includes tree parameters. each tree is output as an integer ID (the order of visitation in the chain, 1,2,...). and listen — make sure to use -i to hide any trees containing lengths/times/rates. fr fr, do not skip that part.
• -P — parallel chains. this implies that several files of (parallel) samples are listed. multi-file moment 🌈

---

🌟 URL 🌟

the prime-phylo home page: http://prime.sbc.su.se

---

💜 SEE ALSO 💜

primeDTLSR(1), primeGSRf(1), primeGEM(1), chainsaw(1), reconcile(1), reroot(1), tree2leafnames(1), treesize(1)

---

and that's the whole page, bestie 🥹 this was prime-phylo 1.0.11, dated 13 Mar, 2012 — yes she's an elder but she still ate 💅 mwah, go analyze something 🎀