Release v0.6.11 - Some new options and bug fixes

Version 0.6.11 (Release on 24 Feb 2026)

• Added option --rename to write clipped bases to the read ID. Works in all modes with options --clip_(r1/r2) and --three_prime_clip_(r1/r2), as well as --hardtrim5 and --hardtrim3. Requested in #166.

• Added option --bgiseq to trim BGISEQ/DNBSEQ/MGISEQ adapters instead of the default auto-detection. Uses AAGTCGGAGGCCAAGCGGTCTTAGGAAGACAA for Read 1 (BGI/MGI forward), and
  AAGTCGGATCGTAGCCATGTCGTTCTGTGAGCCAAGGAGTTG for Read 2 (BGI/MGI reverse). Requested in #196

• Added option --demux <barcode_file> to demultiplex files from a 3'-end barcode after trimming is completed. Requested in #199

• Added option --cutadapt_args "<ARGS>" to pass extra arguments to Cutadapt, enabling use of advanced Cutadapt options without modifying Trim Galore. Please note: not all Cutadapt options are guaranteed to work within Trim Galore.

• Changed Epigenetic Clock processing behaviour (--clock) for the 5' end of Read 2.

• Fixed --demux handling of CR (carriage return) characters in barcode files; fixed barcode length issue with NoCode; added barcode description to demux summary output.

• Fixed RRBS-specific trimming being silently bypassed when --nextseq and --rrbs are used together (#210).

v0.6.10 - add default decompression path

This release fixes the missing default declaration of gzip as decompression path for systems where igzip (and pigz) were not installed.

v0.6.9 - fix declaration bug

This release fixes a declaration bug that crept by merging too many different branches. PRs against the dev branch should fix this in the future.... Here is the commit

Version 0.6.8

Version 0.6.8

• Added new option --stranded_illumina to allow trimming of the adapter sequence ACTGTCTCTTATA (which looks like the Nextera sequence but with an additional A from A-tailing). See also here: #127.

• Trim Galore will now preferentially use igzip for decompression, if installed. More info here

• finally dropped the option --trim1 entirely. It wasn't useful beyond Bowtie 1 paired-end mode, and hence people should cease using it

• the option --max_n COUNT now interprets value between 0 and 1 as fraction of the read length (see here)

• enabled the option --max_length also for paired-end trimming (of small RNAs)

v0.6.7 - DOI via Zenodo

This release is required to generate a new DOI via Zenodo.

v0.6.6

Version 0.6.6

• Changed the way in which we test for the version of Cutadapt, more here:

• Allowed specifying of multiple adapters for special cases. Works either via the command line, e.g.: -a " AGCTCCCG -a TTTCATTATAT -a TTTATTCGGATTTAT" or via a FastA file, like so: -a "file:multiple_adapters.fa" More info here:

• Added new special trimming mode for UMIs for the IMPLICON method (--implicon). In this mode, an 8bp UMI (unique molecular identifier) sequence is transferred from the start of Read 2 to the readID of both reads to allow UMI-aware deduplication (e.g. with deduplicate_bismark --barcode or UmiBam. Following this, Trim Galore will exit.

v0.6.5

Version 0.6.5

• Added checks for whitespace(s) within input filenames, or a potential output folder name (supplied with -o). [FATAL ERROR] messages will advise users to use _ instead.

• In a --paired --basename BASE scenario, the output files will now be called BASE_val_1.fq.gz BASE_val_2.fq.gz as described in the documentation (we previously also added _R1 and _R2). This had to be addressed twice (0f631e5 and 9ad0196) as the first approach was generating the Read 1 twice.

• removed a superflous warning statement for directional RRBS mode

0.6.4

Version 0.6.4

• Changed the adapter auto-detection procedure so that inconclusive detection always defaults to --illumina, unless none of the top 2, equal contaminants was 'Illumina', in which case it now defaults to --nextera. A warning message about this is now printed to the screen as well as to the trimming report.

• Further to this auto-detection precedence behaviour, added the option --consider_already_trimmed INT. If no specific adapter exceeds this limit during the adapter auto-detection, the file is considered 'already adapter-trimmed' and will not be adapter-trimmed again. Quality trimming is carried out as usual (technically, the adapter sequence is set to -a X). This option was added so that pipelines that are being fed either already trimmed or untrimmed data will do the right thing in both cases.

• Changed the trimming mode for paired-end --rrbs in conjunction with --non_directional: previously, Read 2 was only trimmed for CGA or CAA at the 5' end, but not trimmed for read-through contamination at the 3' end if no 5' contamination had been removed. This problem had been introduced in v0.4.3, but since non-directional RRBS is not very common it had not been spotted so far.

• File names for single-end trimming are now changed correctly when both --output_dir and --basename were specified together (was working correctly for PE mode already)

0.6.3

• Also added the number of PolyA trimmed bases to the start of the read in the format trimmed_bases:A:

So an example trimmed read would look like this:

 @READ-ID:1:1102:22039:36996 1:N:0:CCTAATCC
GCCTAAGGAAACAAGTACACTCCACACATGCATAAAGGAAATCAAATGTTATTTTTAAGAAAATGGAAAATAAAAACTTTATAAACACCAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA

@32:A:READ-ID:1:1102:22039:36996_1:N:0:CCTAATCC_PolyA:32
GCCTAAGGAAACAAGTACACTCCACACATGCATAAAGGAAATCAAATGTTATTTTTAAGAAAATGGAAAATAAAAACTTTATAAACACC

0.6.2

• Changed the version checking mechanism so that Trim Galore even works in single-core mode if the version of Cutadapt was 7 years old...

• Fixed setting -j $cores for Cutadapt versions 2.X or above.