Skip to content

Add new module: telogator2#11033

Merged
pinin4fjords merged 12 commits intomasterfrom
new-module/telogator2
Mar 24, 2026
Merged

Add new module: telogator2#11033
pinin4fjords merged 12 commits intomasterfrom
new-module/telogator2

Conversation

@pinin4fjords
Copy link
Member

@pinin4fjords pinin4fjords commented Mar 24, 2026 •
edited
Loading

Summary

  • Add nf-core module for telogator2, a tool for allele-specific telomere length (TL) estimation and telomere variant repeat (TVR) characterization from long-read sequencing data (ONT/PacBio)
  • Uses bioconda package telogator2=2.2.3 with Wave-built Docker and Singularity containers
  • Accepts BAM/CRAM input with optional reference genome, uses topic-based version reporting

Inputs

Input Type Description
reads BAM/CRAM Long-read alignments (PacBio HiFi or ONT)
reads_index BAI/CRAI Index for reads
fasta + fai FASTA + FAI Optional reference genome (single channel)

Outputs

Output Type Description
tlens TSV Telomere lengths per allele (tlens_by_allele.tsv)
plots PNG Allele and violin plots (optional)
qc Directory QC stats, read length distributions, run metadata

Notes

  • The tool exits with code 1 when no telomere reads are found in the input, with a clear error message. This is not caught by the module - pipelines should handle this if mixed samples are expected.
  • Test data uses real ONT telomeric reads from GIAB HG002 (Add ONT telomere test data (GIAB HG002) test-datasets#1947), plus a PacBio BAM without telomere reads to test the failure path.
  • TL values are non-deterministic across runs (minimap2 alignment), so tests assert the TSV header structure and output existence rather than exact md5 checksums. A fixed random seed (--rng 42) is set in the test config.

Test plan

  • nf-test passes for all 3 tests (telomere reads, no telomere reads, stub) with --profile docker
  • Checked stderr for silent errors - none found
  • CI nf-test validation

🤖 Generated with Claude Code

@pinin4fjords @claude
Add new module: telogator2
7649cbb 
Add nf-core module for telogator2, a tool for allele-specific telomere
length estimation and TVR characterization from long-read sequencing
data (ONT/PacBio).

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
@pinin4fjords pinin4fjords marked this pull request as draft March 24, 2026 11:26
@pinin4fjords
Copy link
Member Author

pinin4fjords commented Mar 24, 2026 •
edited
Loading

TODO: Generate telomere-containing long-read test data so the tests exercise telogator2's actual analysis path, not just the "no telomere reads found" fallback.

@pinin4fjords pinin4fjords mentioned this pull request Mar 24, 2026
Merged
pinin4fjords and others added 8 commits March 24, 2026 11:45
@pinin4fjords @claude
add telomere test data and temporarily point to fork
f650efa 
- Add ONT telomere reads test (exercises real analysis path)
- Keep PacBio no-telomere test (exercises graceful fallback)
- Temporarily override modules_testdata_base_path to
  pinin4fjords/test-datasets#telogator2-test-data (revert before merge)

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
@pinin4fjords
Merge branch 'master' into new-module/telogator2
90e6a2b
@pinin4fjords @claude
fix: combine fasta and fai into single reference channel
0cccf38 
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
@pinin4fjords @claude
fix: simplify process script and assert failure on no telomere reads
cd0900e 
Remove error-catching wrapper from telogator2 process script. When no
telomere reads are found the tool now fails with a clear error message,
which the no-telomere test asserts against.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
@pinin4fjords
Merge branch 'new-module/telogator2' of github.com:nf-core/modules in…
de72854 
…to new-module/telogator2
@pinin4fjords @claude
refactor: emit individual files instead of directory
affae4a 
- tlens: tlens_by_allele.tsv (primary result)
- plots: *.png (allele and violin plots, optional)
- qc: qc directory (stats, read lengths, metadata)

Also revert modules_testdata_base_path now that
nf-core/test-datasets#1947 is merged.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
@pinin4fjords @claude
fix: exclude non-deterministic rng.txt from snapshot
e0be595 
The qc/rng.txt file contains a random seed that differs across runs.
Assert qc output exists but don't snapshot it.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
@pinin4fjords @claude
fix: handle non-deterministic telogator2 outputs in tests
09f3ff3 
Set fixed random seed (--rng 42) via test config. Assert tlens header
structure rather than md5 since TL values vary across runs due to
minimap2 non-determinism. Assert plots and qc exist without
snapshotting.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
@pinin4fjords pinin4fjords marked this pull request as ready for review March 24, 2026 12:42
fellen31
fellen31 reviewed Mar 24, 2026
View reviewed changes
Copy link
Contributor

@fellen31 fellen31 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Nice!

pinin4fjords and others added 2 commits March 24, 2026 13:22
@pinin4fjords @claude
fix: address PR review feedback
c95462f 
- Use module_args pattern for ext.args in test config
- Snapshot output file names (not md5s) for non-deterministic outputs
- Remove PNGs from stub (plots are optional)

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
@pinin4fjords @claude
fix: make plots a required output
601b8c4 
The two main plots (all_final_alleles.png, violin_atl.png) are always
produced on a successful run. Remove optional flag and add them back
to the stub.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
@pinin4fjords pinin4fjords requested a review from fellen31 March 24, 2026 13:30
fellen31
fellen31 reviewed Mar 24, 2026
View reviewed changes
Copy link
Contributor

@fellen31 fellen31 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I would just touch all QC files in the stub section, so that they are output correctly in a pipeline stub run. Otherwise, everything looks good!

@pinin4fjords @claude
refactor: split QC directory into individual output channels
09b23d0 
Emit cmd, stats, qc_readlens, readlens, and rng as separate channels
instead of a single qc directory. Touch all QC files in stub.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
@pinin4fjords pinin4fjords requested a review from fellen31 March 24, 2026 13:45
fellen31
fellen31 approved these changes Mar 24, 2026
View reviewed changes
Copy link
Contributor

@fellen31 fellen31 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks good to me!

@pinin4fjords pinin4fjords added this pull request to the merge queue Mar 24, 2026
Merged via the queue into master with commit 690bb56 Mar 24, 2026
25 checks passed
@pinin4fjords pinin4fjords deleted the new-module/telogator2 branch March 24, 2026 13:59
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@fellen31 fellen31 fellen31 approved these changes

Assignees

No one assigned

Labels

size/m

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@pinin4fjords @fellen31