MAC3_BDGCMP(1)

💖 MAC3_BDGCMP(1) 💖

okay bestie, gather round, because today we're talking about mac3_bdgcmp — Model-based Analysis for ChIP-Sequencing 🧬✨ (yes, it's the bdgcmp command you run as macs3 bdgcmp, she's iconic, she compares your two bedGraphs and gives you a score, no cap)

---

✨ NAME ✨

mac3_bdgcmp — Model-based Analysis for ChIP-Sequencing 🎀

---

🌸 DESCRIPTION 🌸

so like, here's the whole vibe, this is literally the usage line, fr fr:

usage: macs3 bdgcmp [-h] -t TFILE -c CFILE [-S SFACTOR] [-p PSEUDOCOUNT]

[-m {ppois,qpois,subtract,logFE,FE,logLR,slogLR,max}
[{ppois,qpois,subtract,logFE,FE,logLR,max} ...]]
       [--verbose VERBOSE] [--outdir OUTDIR] (--o-prefix OPREFIX | -o
       OFILE [OFILE ...])

like, that's the shape of it 💅

---

🦋 options 🦋

• -h, --help — shows this help message and then peaces out 👋
• -t, --tfile TFILE — the Treatment bedGraph file, like *_treat_pileup.bdg from MACSv2. REQUIRED, bestie, you literally cannot skip this one 🔥
• -c, --cfile CFILE — the Control bedGraph file, like *_control_lambda.bdg from MACSv2. Also REQUIRED, no escaping it 💀
• -S, --scaling-factor SFACTOR — scaling factor for the treatment and control track. keep it at 1.0 or just leave the default in most cases, okay? only set it while you have SPMR output from MACS3 callpeak, and you're planning to calculate scores the way the MACS3 callpeak module does. if you wanna simulate 'callpeak' w/o '--to-large', calculate the effective smaller sample size after filtering redundant reads, in millions (like, put 31.415926 if the effective reads are 31,415,926) and feed that to '-S'; for 'callpeak --to-large', calculate the effective reads in the larger sample instead. DEFAULT: 1.0 ⭐
• -p, --pseudocount PSEUDOCOUNT — the pseudocount used for calculating logLR, logFE or FE. the count gets applied after the normalization of sequencing depth, just so you know. DEFAULT: 0.0, no pseudocount is applied 🍓
• -m, --method {ppois,qpois,subtract,logFE,FE,logLR,slogLR,max} [{ppois,qpois,subtract,logFE,FE,logLR,max} ...] — the method to use while calculating a score in any bin by comparing the treatment value and the control value. the available choices are: ppois, qpois, subtract, logFE, FE, logLR, slogLR, and max. and what do they mean, you ask? let me spell it out 💅
  • ppois = Poisson Pvalue (in -log10(pvalue) form), using control as lambda and treatment as observation
  • qpois = q-value through a BH process for poisson pvalues
  • subtract = subtraction from treatment
  • logFE = log10 fold enrichment (you need to set pseudocount for this one!)
  • FE = linear scale fold enrichment
  • logLR = log10 likelihood between ChIP-enriched model and open chromatin model (also needs pseudocount set!)
  • slogLR = symmetric log10 likelihood between two ChIP-enrichment models
  • max = maximum value between the two tracks Default option is ppois. periodt 👑
• --verbose VERBOSE — sets the verbose level of runtime messages. 0: only show critical message, 1: show additional warning message, 2: show process information, 3: show debug messages. DEFAULT: 2 🌈
• --outdir OUTDIR — if specified, all output files get written to that directory instead. default: the current working directory 📁
• --o-prefix OPREFIX — the PREFIX of the output bedGraph file to write scores into. if you give it as A, and the method is 'ppois', the output file will be A_ppois.bdg. mutually exclusive with -o/--ofile — you can't have both, sorry babe 🙈
• -o, --ofile OFILE [OFILE ...] — output filename. mutually exclusive with --o-prefix. and heads up: the number and the order of arguments for --ofile must be the same as for -m. the way that it has to match exactly… 😭

---

lowkey reminder: --o-prefix and -o/--ofile are mutually exclusive, so pick one and commit, main character energy only 💅

---

and that's the whole thing, bestie, you're so ready 🫶 this lil manual is from macs3 bdgcmp 3.0.4, dated May 2026 — slay, go compare those tracks ✨👑