Release notes: Latest releases

See older releases for earlier versions.

Version 1.8.2

Release date: 02.12.2019

All changes

  • Fixed bug where quality filtering did not work for multiple probands.
  • Fixed bug where importing the same sample more than once caused a crash.
  • Fixed bug causing several variants not to be filtered out as they should in the region filter.

Version 1.8.1

Release date: 30.10.2019

All changes

  • Fixed bug where users were not able to finalize analyses with reused allele assessments and auto-ignored references under "Pending evaluation".
  • Fixed bug where reference data was not reloaded when including a variant from filtered variants.

Version 1.8

Release date: 30.10.2019

Highlights

Add "other" and unweighted ACMG criteria

Sometimes, criteria that don't match the ACMG guidelines are important for a variant interpretation, e.g. the ENIGMA criteria for the BRCA1/BRCA2 genes. ELLA now allows adding these to the interpretation as a generic OTHER criterion. The type and impact on the classification should be given in this criterion's comment field once added.

In addition, users can often spend significant time evaluating an ACMG criterion for a particular interpretation, but in the end decide that the requirements are not met. ELLA now allows setting an added ACMG criterion as NOT WEIGHTED, so that comments related to this work can be properly recorded.


Figure: The new "other" criterion and unweighted option for ACMG criteria.

Note that neither "other" or unweighted ACMG criteria are counted in the calculation of suggested classification.

Filter improvements: Gene and allele ratio

The filter settings now allows using genes as a variable in rules for filters or exceptions. This allows conditioning any rule on the presence/absence of a gene, e.g. exclude certain genes from a particular filter.

In addition, it is now also possible to use allele ratio (alternative allele reads/total reads) as a variable in the quality filter. In our experience, this gives a more powerful filter than using the qual variable, especially for high sequencing depths. A caveat is that mosaic variants may be missed, but this can be partially be circumvented by adding particular genes where mosaics are expected to a gene exclusion rule as described above.

All changes

Last Updated: 12/2/2019, 1:10:50 PM