Add binner SemiBin2

[d4straub]

Adds SemiBin2, related to #874

SemiBin2 handles cases where it finds no bins gracefully: exit code 0. It does error though when there are too less contigs (but I could not find how many are required). Therefore errors dont need to be ignored, I think.

Whenever long reads are used in the assembly (if meta.lr_platform) then long read binning mode is activated. However I am not sure that is supposed to be used for SPAdes assemblies, because they are (at least last time I checked) short read assembles with long read stitching. But I follow the FAQ What should I do if I have hybrid data (short- and long-reads)? From SemiBin's point-of-view, you should generally treat this using the long-reads pipeline (--sequencing-type=long_read). from here.

About CI tests:

• 8/8 succeed: 3 with SemiBin2 activated; see post below
• test_alternatives should work as well and could be activated (requires snap update though)
• longreadonly & longreadonly_alternatives produce assemblies with only 3 contigs, SemiBin2 complains; couldnt find any required minimum contig number though.

Still to do:

• confirmation for successful -profile test_full pending (test running atm)

PR checklist

• This comment contains a description of changes (with reason).
• If you've fixed a bug or added code that should be tested, add tests!
• If you've added a new tool - have you followed the pipeline conventions in the contribution docs
• If necessary, also make a PR on the nf-core/mag branch on the nf-core/test-datasets repository.
• Make sure your code lints (nf-core pipelines lint).
• Ensure the test suite passes (nextflow run . -profile test,docker --outdir <OUTDIR>).
• Check for unexpected warnings in debug mode (nextflow run . -profile debug,test,docker --outdir <OUTDIR>).
• Usage Documentation in docs/usage.md is updated.
• Output Documentation in docs/output.md is updated.
• CHANGELOG.md is updated.
• README.md is updated (including new tool citations and authors/contributors).

[d4straub]

@nf-core-bot fix linting

[d4straub]

Investigated CI test outcomes:

• 1 '-profile test' -> apparently no new bin files (locally SemiBin does produce bins), summary files changed -> snap mismatch causes fail
• [skipped] 2 '-profile test_alternatives' has warning:

WARN: Input tuple does not match tuple declaration in process `NFCORE_MAG:MAG:DOMAIN_CLASSIFICATION:TIARA:TIARA_SUMMARY` -- offending value: [[:], [/home/runner/_work/mag/mag/~/tests/c090c86a5a5a2ac52952ba3cc887e719/work/e0/bf0d5166fd4ce157c46c583edeb3c7/MEGAHIT-SemiBin2-bins-test_minigut.binclassification.tsv], [:], [/home/runner/_work/mag/mag/~/tests/c090c86a5a5a2ac52952ba3cc887e719/work/d4/b3e898cef3fa26b001a9a0d38099dc/MEGAHIT-MetaBAT2-bins-test_minigut.binclassification.tsv]]

-> TIARA was not executed
-> new bin files, summary files changed -> snap mismatch causes fail
-> locally I can confirm the warning, but the pipeline executes all expected processes

• 3 '-profile assembly_input' new bin files, summary files not changed -> snap mismatch causes fail
• 4 '-profile hybrid' no new bin files! but process was executed and summary files changed -> snap mismatch causes fail
• [skipped: assembly with too less contigs] 5 '-profile longreadonly' fails with:

Process `NFCORE_MAG:MAG:BINNING:SEMIBIN_SINGLEEASYBIN (minigut)` terminated with an error exit status (1)
2025-11-14 14:46:41 99ff53988499 SemiBin2[32] ERROR There are 3 contigs in input file FLYE-minigut.assembly.fasta, but only 3 contain(s) at least 1500 basepairs.

• [skipped: assembly with too less contigs] 6 '-profile longreadonly_alternatives' fails as above
• [skipped: because minimal test] 7 '-profile test_minimal'
• [skipped: because too short contigs] 8 '-profile test_single_end' fails with:

Process `NFCORE_MAG:MAG:BINNING:SEMIBIN_SINGLEEASYBIN (test_minigut_sample2)` terminated with an error exit status (1)
2025-11-14 14:44:43 bf456a6a163f SemiBin2[32] ERROR There are 215 contigs in input file MEGAHIT-test_minigut_sample2.fa, but only 1 contain(s) at least 1500 basepairs.

--> skip that because data isnt appropriate.

Conclusion: Most tests might be fine, but not long reads only tests. I'll see what I can do to make them pass.

[dialvarezs]

Looks very clean!

Minor comments from my end.

[d4straub]

Profile test_full successfully finished. I think though SemiBin2 and COMEBin need relatively long, maybe we have to make them opt-in in future.

Thanks for all the comments and the reviews!