MCALLER.PY(1)

💖 MCALLER.PY(1) 💖

okay bestie, sit down, hydrate, and let me tell you about mCaller.py — that's the manual page for the User Commands section, and it's literally the moment. no cap.

🌸 NAME 🌸

mCaller.py — find methylation in nanopore reads. yes. that's the whole job. iconic behavior fr fr.

✨ DESCRIPTION ✨

okay so first things first, sometimes it comes out the gate being dramatic:

/usr/lib/python3/dist-packages/joblib/_multiprocessing_helpers.py:53:
UserWarning: [Errno 13] Permission denied.  joblib will operate
in serial mode

and then it's like:

warnings.warn('%s.
       joblib will operate in serial mode' % (e,))

that's a UserWarning about [Errno 13] Permission denied, and joblib will operate in serial mode. just so you know what you're walking into 💀

the usage situation is:

usage: mCaller [-h] (-p POSITIONS | -m MOTIF) -r REFERENCE -e TSV -f
FASTQ

       [-t THREADS] [-b BASE] [-n NUM_VARIABLES] [--train]
       [--training_tsv TRAINING_TSV] [-d MODELFILE] [-s SKIP_THRESH] [-q
       QUAL_THRESH] [-c CLASSIFIER] [--plot_training] [-v]

and what does it do, you ask? Classify bases as methylated or unmethylated. that's the vibe. that's the whole serve. 🧬✨

🎀 optional arguments 🎀

• -h, --help — show this help message and exit. classic. the "help me bestie" button 🫶
• -p POSITIONS, --positions POSITIONS — file with a list of positions at which to classify bases. it must be formatted as a space- or tab-separated file with chromosome, position, strand, and label if training. yes, all four, don't skip the label if you're training, she's watching 👀
• -m MOTIF, --motif MOTIF — classify every base of type --base in the motif specified instead. can be a single one-mer, so like, one little guy. small but mighty 💅
• -r REFERENCE, --reference REFERENCE — fasta file with reference aligned to.
• -e TSV, --tsv TSV — tsv file with nanopolish event alignment.
• -f FASTQ, --fastq FASTQ — fastq file with nanopore reads.
• -t THREADS, --threads THREADS — specify number of processes (default = 1). one. singular. alone. give her friends, bestie 🦋
• -b BASE, --base BASE — bases to classify as methylated or unmethylated (A or C, default A).
• -n NUM_VARIABLES, --num_variables NUM_VARIABLES — change the length of the context used to classify. default of 6 variables corresponds to an 11-mer context (6*2-1). do the math, it's giving arithmetic 💜
• --train — train a new model. requires labels in positions file. no labels, no training, that's the rules ✋
• --training_tsv TRAINING_TSV — mCaller output file for training.
• -d MODELFILE, --modelfile MODELFILE — model file name.
• -s SKIP_THRESH, --skip_thresh SKIP_THRESH — number of skips to allow within an observation (default 0).
• -q QUAL_THRESH, --qual_thresh QUAL_THRESH — quality threshold for reads (default none). none. free. unbothered ✨
• -c CLASSIFIER, --classifier CLASSIFIER — use alternative classifier: options = NN (default), RF, LR, or NBC. non-default may significantly increase runtime, so like, be aware, she's slow when you switch it up 😭
• --plot_training — plot probabilities distributions for training positions. requires labels in positions file and --train. both of them. all of them. periodt 📈
• -v, --version — print version.

and that's the whole page bestie, you ate that 🍓💖

before you go — this gorgeous thing is mCaller.py 1.0.3+git20210331.f7... and it's dated June 2021, sitting pretty under MCALLER.PY(1) in the User Commands manual. mwah. go classify some bases 👑🌈