Hard-trimming

--hardtrim5 and --hardtrim3 are run-and-exit modes that hard-clip every read to a fixed length, then exit. They run before quality and adapter trimming and don't touch report output.

Hard-trimming to leave bases at the 5' end

--hardtrim5 INT clips reads down to their first INT bases. Output goes to {stem}.{INT}bp_5prime.fq(.gz). Useful for shortening reads to a uniform length:

before:         CCTAAGGAAACAAGTACACTCCACACATGCATAAAGGAAATCAAATGTTATTTTTAAGAAAATGGAAAAT
--hardtrim5 20: CCTAAGGAAACAAGTACACT

Hard-trimming to leave bases at the 3' end

--hardtrim3 INT keeps only the last INT bases. Output goes to {stem}.{INT}bp_3prime.fq(.gz). Useful for removing a fixed bias at the 5' end (e.g. UMIs or inline barcodes):

before:         CCTAAGGAAACAAGTACACTCCACACATGCATAAAGGAAATCAAATGTTATTTTTAAGAAAATGGAAAAT
--hardtrim3 20:                                                   TTTTTAAGAAAATGGAAAAT

When to use this

Hard-trimming is a sledgehammer: quality is ignored, adapters are ignored, every read is clipped to the same length. It's the right tool for:

• Standardising read lengths for tools that require fixed-length input.
• Removing a known fixed bias at one end (e.g. UMI or inline barcode) before passing data through a second Trim Galore run for adapter and quality trimming.
• Producing equally-sized inputs for benchmarking.

For everything else, prefer the fine-grained --clip_R1, --clip_R2, --three_prime_clip_R1, --three_prime_clip_R2 flags, which apply on top of normal quality and adapter trimming.

Compatibility

Both modes accept multiple input files in a single invocation and process them sequentially. Paired-end mode is not required. --hardtrim* operates per-file independently.