SVS 8.5.0 Release Notes
- The GBLUP and K-Fold cross validation methods have been extended with a “Large-N” mode allowing them to scale to samples sizes (over 8,000) that would have previously been prohibited due to the exponential scaling of the memory requirements of the exact method. See Large Kinship Matrices or Large Numbers of Samples.
- Fixed the human GRCh_38 assembly so each chromosome had the correct length specified.
- Fixed issues with computing lengths of chromosomes when converting FASTA data to create a new genome assembly.
- Fixed haplotype block spreadsheet selection dialog logic to only activate haplotype block spreadsheets.
- Fixed bug that caused crash when plotting numeric plots in GenomeBrowse with inactivated columns.
- VCF export now correctly handling insertions and deletions in scaffold chromosomes.
- Because K-fold Cross Validation for Genomic Prediction changes the row-state as it creates row subset, it now automatically subsets a spreadsheet with inactivated rows.
- Fixed issue with manually specifying and saving affection status during VCF import on Mac OSX.
- Added deprecated flags to the Data Source Library so older annotation sources could be hidden automatically when the current box is checked.
- Added parameter to get the amount of RAM allocated as the working memory size, ghi.const.MemoryWorkSize.
- Improved speed of adding a large number of files to the VCF importer.