JBrowse 1.16.5
Minor improvements
-
Added better error reporting if there is a BAI file for a genome that
is longer than 2^29 and requires a CSI index -
Added default dontRedispatch=chromosome,region which avoids long
GFF3Tabix redispatching for common gff3 types that don't need it. -
Added a
hideIframeDialogUrl
option to hide the iframe url in a
iframeDialog
popup. Thanks to @mictadlo for the suggestion (issue
#1365) -
Updated @gmod/tabix and @gmod/bam to unzip bgzf chunks as they are
received instead of performing chunk merges. This also allows
actual fileOffset to be used as a feature ID instead of CRC32 hash
(@cmdcolin) -
Added regularization of chromosome names using roman numerals,
common in both S. cerevisiae and C. elegans genome communities.
Thanks to @scottcain for assistance (pull #1376, @cmdcolin) -
Added ability for vertical scroll events to bubble out of the jbrowse
div or iframe, which can be more intuitive in embedded jbrowse's.
Otherwise, the GenomeView stopped all vertical scroll events in it's
area. If you need the old behavior usealwaysStopScrollBubble
.
(@cmdcolin, pull #1373)
Bug fixes
-
Fixed a long standing issue with BigWig files not accessing the
lowest summary zoom level (e.g. the first one after unzoomed).
(@cmdcolin, issue #1359) -
Fixed a bug with using the scoreType: maxScore configuration on
Wiggle tracks failing at non-summary zoom levels. Thanks to
@patrickcng90 for reporting -
Fixed a bug with name regularizing that particularly affected CRAM
on mitochondrial chromosomes. Thanks to @sachalau for patience and
sample data for debugging this issue! (issue #1367, @cmdcolin) -
Fixed issue with GFF3/Tabix types colliding their attribute names
with the field e.g. if start=0 was in column 9. Thanks to
@loraine-gueguen for reporting (issue #1364). -
Fixed an issue with NeatHTMLFeatures when zoomed out. Thanks to
@abretaud for the implementation and fix (pull #1368). -
Restored ability to access widest zoom level from bigwig. Thanks
to @lukaw3d for noticing and debugging (issue #1375, @cmdcolin) -
Fixed some minor parsing of 'b' type data series in CRAM files,
which happens when CRAM is not using reference based compression
so it is uncommon (@cmdcolin) -
Added some more checks for track types for combination tracks (issue
#1361)