GENRANDOMREADS(1)

✨ NAME ✨

genRandomReads — a whole toolkit for processing next-gen sequencing data, and honestly it's giving main character energy 💅

---

💖 DESCRIPTION 💖

okay bestie, first things first: if you forget stuff, she WILL yell at you. like literally, these are the exact complaints you might get 👇

• Error: a transcript file must be provide.
• Error: the output prefix must be provide.
• Error: the wanted expression levels must be provided.
• Warning: no read number is specified. Generating one million reads.

(not the grammar being a lil unhinged, but we move 🫶)

🌸 Usage 🌸

For scanning a FASTA/gz file:

/usr/lib/subread/genRandomReads --summarizeFasta \

--transcriptFasta <file> --outputPrefix <string> [--simpleTranscriptId]

For generating read/pairs:

/usr/lib/subread/genRandomReads --transcriptFasta <file>\

--outputPrefix <string> --expressionLevels <file> [other options]

---

🎀 The Options, one by one, no skips 🎀

• --summarizeFasta — she only spits out the transcript names and their lengths. that's it, that's the whole job. lowkey iconic.
• --transcriptFasta <file> — your transcript database, in FASTA/gz format. the source material fr fr.
• --outputPrefix <string> — the prefix that gets slapped onto the output files. name your babies ✨
• --totalReads <int> — total read/pairs in the output.
• --expressionLevels <file> — a two-column table, delimited by <TAB>, holding the TPM values you want. the columns are: TranscriptID and TPM. yes you need both, no you can't skip the tab 🙈
• --readLen <int> — how long the output reads are. default is 100, in case you were wondering.
• --totalReads <int> — total read/pairs in the output. (yes, she shows up twice in the list, and we respect her for it 👑)
• --randSeed <int64> — the seed for generating random numbers. by default UNIXTIME is used as the random seed, so like, time itself is the vibe check.
• --qualityRefFile <file> — a textual file full of Phred+33 quality strings, used for simulating sequencing errors. the quality strings have to be the same length as the output reads, no cap. if you omit this option, no sequencing errors get simulated at all. simple as.
• --floorStrategy — how to deal with round-up errors, and you've got three moods:
  • 'FLOOR': generate fewer reads than wanted
  • 'RANDOM': randomly assign the margin reads to transcripts
  • 'ITERATIVE': hunt down the best M value to get ~N reads
• --pairedEnd — generate paired-end reads. two for the price of one, periodt.
• --insertionLenMean <float>, --insertionLenSigma <float>, --insertionLenMin <int>, --insertionLenMax <int> — the parameters of a truncated normal distribution that decides the insertion lengths of paired-end reads. the default values are: mean=160, sigma=30, min=110, max=400.
• --simpleTranscriptId — truncate transcript names down to the first '|' or space. chop chop ✂️
• --truthInReadNames — encode the true locations of the reads inside the read names. snitch behaviour, but make it science 🧚
• --noActualReads — do NOT actually generate reads in the fastq.gz files. all vibes, zero output files, truly a state of mind.

---

👑 AUTHOR 👑

This manpage was written by Alexandre Mestiashvili for the Debian distribution, and it can be used for any other usage of the program. shoutout to them, genuinely 💜

---

and that's a wrap bestie!! this was genRandomReads 2.0.3, dated March 2023, so if you're reading this way later... still slay, still the same version number 🌈✨ bye ily, go summarize some fasta 🦋