Skip to content

Shrink Reads

Summary

Subset FASTQ reads by count or fraction for lightweight test datasets.

Description

Supports deterministic head-style subsampling (first_n) and random sampling (random) for single-end, interleaved, and paired-end layouts.

This command is intended for quick dry runs and resource-reduced tests, not as a full read-normalization strategy.

Usage

rolypoly shrink-reads [OPTIONS]

Options

  • -i, -in, --input: Input raw reads file(s) or directory containing them. For paired-end reads, you can provide an interleaved file or the R1 and R2 files separated by comma. If a directory is provided, one output per input identified file/pair will be created. (type: TEXT; default: Sentinel.UNSET)
  • -st, --subset-type: how to sample reads from input. (type: CHOICE; default: top_reads)
  • -sz, --sample-size: For top_reads/random, at most this many reads (or proportion if <1). For bbnorm, this is the target k-mer depth. (type: FLOAT; default: 1000)
  • --bbnorm-min-depth: Minimum depth threshold for bbnorm normalization (min in bbnorm.sh). (type: INTEGER; default: 2)
  • -t, --threads: Number of worker threads. (type: INTEGER RANGE; default: 1)
  • -g, --log-file: Path to the log file. (type: FILE; default: rolypoly.log)