v2.0.0

[nebfield]

We've marked this release as the first full release of v2, linked with our recent publication describing the calculator in full (Lambert, Wingfield, et al. Nature Genetics. 2024.).

Changelog

Improvements

• Make report shareable by default
  • Remove individual level data
  • Don't show density plots with small sample sizes
• Add warnings about complex alleles (e.g. HLA/APOE) and dosage specific effect weights to the report
• The variant verification step added in 2.0.0-beta.3 has been integrated into pgscatalog-aggregate
  • The symmetric difference of scoring file variant IDs (.scorefile.gz) and variants that contributed to the final calculated score (.vars plink file) must be an empty set

Bug fixes

• Stop crashing when encountering a scoring file with dosage specific effect weights (skip instead)
• Fix report logo
• Fix VCF input with JSON samplesheets
• Add tar to zstd conda environment to prevent very old tar installs failing to extract the database
• Reduce download max thread workers to prevent throttling by the EBI
• Fix variant verification step failing in some conda deployments

Full Changelog: v2.0.0-beta.3...v2.0.0

---

Note

• The COMBINE_SCOREFILES process may take longer to finish and use more memory than in previous versions
• New internal variant data models were added in this release to improve handling complex alleles and dosage specific effect weights
  • Also, every variant (scoring file row) has many more validation steps now to ensure data quality and consistency
• Speed and memory usage will be improved in the next release

---

This discussion was created from the release v2.0.0.