v365

Code changes:

• Fixed silent failure in custom tracks when incomplete installation, loaders missing (#21330). Hiram
• Improved performance in custom track loading replace hTableExists() with sqlTableExists() when db==customTrash (#21330). Hiram
• Update Short Match: don't do TATAWARR by default, slow for large regions (#20969). Angie
• Enhance hgTables: support hub track names with '.', GTF output for hub tracks (#21353). Angie
• Update hub trackDb: errAbort with useful message instead of assert for missing 'view' (#21386). Angie
• Enhance gff3ToRefLink.pl: fix missing data in ncbiRefSeqLink. Angie
• Add small tweaks to build & dbSnp scripts to support test database for patches (#18853). Angie
• Fix up openstack installation scripts to use a command line rsa-key pair name instead of a hardcoded name. Christopher
• Fix the right-click track configuration menus in hgTracks, should work again in IE11 (#21196). Jonathan
• Switching to bowtie as the main aligner for CIRM (#21391). Max
• Fixed problem in hgCollection that caused collection track names to get longer and longer (#21304). Brian
• Fixes to interact track type (on RR, not announced or even doc'ed), responding to QA (#21109). Kate
• New GTEx Body Map HTML page (gtexBodyMap.html), ready for QA (#21295). Kate
• Makefile cleanup including several required CIRM tagStorm utilities, removing restriction on CIRM that one must be logged in to use Download All (#21315). Galt
• Allow error-handling (stop on error) for cirm-01 like we have for hgwdev and now make will behave more reliably with warnings and errors. Galt
• Making CIRM pipeline scanner cdwFindSymlinkable handle resubmitted files better. Galt

Data changes:

• Final steps in mm10 ucsc gene build (#19786). Brian
• Work on uniProt, proteome, and go databases (#21347). Brian
• Added snp151 tracks for hg19, hg38 (#21010). Angie
• New sheep/oviAri4 assembly (#17247). Christopher
• Work Towards a NCBI PheGenI track (#21052). Max