For our mid-summer EPI2ME 25.07-01 release, we provide a key update to our human variation workflow, along with several minor enhancements across the broader workflow portfolio. We also introduce a new adaptive sampling based PGx dataset.

wf-human-variation [v2.7.2]
- Updates the bundled Sniffles software to Sniffles v2.6.3. This version of Sniffles incorporates an Oxford Nanopore patch to improve memory management during variant calling that had resulted in silent workflow failures. All users of wf-human-variation v2.7.0 must update to this version and we recommend that users of wf-human-variation v2.7.1 update. There are no changes to the VCF results produced in successful runs. Please review the wf-human-variation changelog for a more detailed explanation.
- The HTML reports produced by the workflow have been update to harmonise the presentation style with other EPI2ME workflow reports.
- A fix has been introduced to ensure that appropriate Clair3 models are used during the small variant calling analyses. A SUP model was previously incorrectly used for the dna_r10.4.1_e8.2_400bps_hac@v5.2.0 and dna_r10.4.1_e8.2_260bps_hac@v4.0.0 Dorado models.
- Updated handling of fatal memory errors ("Error: 137")has been introduced for the pileup_variants and evaluate_candidates steps of the workflow.
- When using the EPI2ME Desktop Application’s IGV functionality the BAM and bedMethyl tracks are now presented.
- A "Regions below target coverage" section has been added to the report that presents BED format information of genomic regions and their depth of coverage based on the alignment_report_coverage_threshold parameter.
wf-teloseq [v1.0.1]
- Patch release providing several bug fixes which have been raised by the community. This release does not affect the determination of telomere repeat boundary location.
- KDE calculations previously crashed the workflow when only a single read passed filtering. This issue has now been handled, and KDE plots will only display when two or more reads pass filtering and have different lengths.
- Sample plots were previously linked to incorrect barcodes in the ‘Read summary plots’ dropdown selector of the report. This issue has been fixed, and the plots are now correctly associated with their respective sample aliases. This fix does not alter any report plotting.

We are pleased to release a new dataset from eight Coriell cell lines sequenced on PromethION using adaptive sampling to enrich 375 pharmacogenes. This dataset supports evaluation of our PGx sequencing end-to-end workflow and is available via the data release page. The dataset download links and instructions for how to access the ont-open-data resources are provided through the data release page.