- 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.
- Added deprecated flags to the Data Source Library so older annotation sources could be hidden automatically when the current box is checked.
- Fixed Y-axis zoom in GenomeBrowse when not in automatic mode.
- Fixed searching indexed fields when annotation sources are loaded from Public Repository.
- Make sure that all executables have the correct permissions for Linux x64 and RHEL builds (aria2c, assistant, etc.)
- For BAM alignment plots, remember the edited value for Filter Multi-Mapped Alignments when the option is checked and unchecked.
- Adjusted Data Source Library Ctrl+A / Ctrl+Shift+A behavior to modify only items in the current view.
- Converted movable toolbars to fixed “flow” toolbars for and adjusted the order of some toolbar items.
- On Lin64, renamed libstc++.so.6 to libstdc++.so.6.rename-if-needed so that GenomeBrowse will run on Ubuntu 15.04.
- When visualizing annotation sources in GenomeBrowse set labels in the following preferred order: “Identifier” > “Ref/Alt” > “Gene Name” > “Name”
- Allow indexing of string array fields for annotation sources. This supports querying against these fields in GenomeBrowse.
- Added HTML format flags into annotation Source Editor so visualization of these fields can be improved though HTML formating.
- Allow for coverage computation for VCF files in GenomeBrowse that do not contain a Genotype (GT) field but does have other sample level FORMAT files.
- Renamed Annotation Download Window buttons to make it clear that downloaded tracks will not be deleted through this dialog.
- Optimized drawing of Manhattan plots to improve speed of drawing and redrawing at the whole genome scale.
- Save as Image now includes the option for saving in SVG format.
- Added support for managing Genome Assembly views in the GenomeBrowse window, genome segment order and visibility can now be changed. See Managing Genome Assembly Views for further details.
- Clicking on a folder URL now launches a file explorer window at that location instead of trying to launch and failing.
- Fixed console information for categorical array fields to have the correct size instead of a list of missing values.
- In the Expression Editor, the “Chr” field is now correctly handled as a string and the “Start” and “Stop” fields are correctly handled as integers. This will now allow you to add expressions such as “Chromosome”, “Start”, “Stop”, “Stop - Start <= 2”, etc.
- Fixed sorting of fields in the Data Source Library to be in numeric order for certain fields including the size of the sources.
- Plots now correctly maintain y-axis zoom levels when selecting Save As Image or when adjusting the x-axis position.
- Fixed index assignment operator for arrays so Expression Editor can use specific entries to calculate new fields.
- Changed SSL (libssl.so.10 and libcrypto.so.10) bundled with RHEL build to prevent SSL Handshake errors when logging in.
- Convert Wizard bugs fixed:
- Fixed error when a CSV file was selected for converting that caused the comma delimiter to be treated as part of the field data.
- Merging TSF files now correctly merges the segment list.
- Fixed an issue that caused a crash when selecting an invalid source for conversion.
- Allow GenomeBrowse to plot data from a VCF file that contains string arrays instead of giving the error that mapFields cannot be converted from StringArray to String.
Added Select All and Deselect All keyboard support (<Ctrl> + A and <Ctrl> + <Shift> + A respectively) to all web views.
Added search pop-up and keyboard search support (<Ctrl> + F) for all web views.
Added Variant Databases as a default location in the Data Source Library under Local Annotations if it exists.
Fixed right-click Open Folder on location sub-folders in the Data Source Library.
Adjusted layout of Add Location dialog in the Data Source Library to prevent the dialog from being too small for its contents.
Updated default RefSeq gene annotation source based on GRCh37_g1k build to include Locus Reference Genome (LRG) identifiers, updated source name is RefSeq Genes 105v2, NCBI.
Convert Wizard polishes:
- Added support for converting BED files with only chromosome, start position, and stop position.
- Added support for vcf.gz (gzip) files.
- Added the ability to copy source information from an existing annotation source to a new source through the Advanced Options.
Registering for GenomeBrowse now has “Keep me informed!” unchecked by default and the tool tip has been updated to state:
“Stay updated with exclusive e-books, timely invitations to webcasts and events and other communications from Golden Helix.”
In the Data Source Library, using the keyboard arrows to select a source will now update the information about that source in the Information Pane.
Clear completed downloads from the download manager after closing GenomeBrowse.
Allow multiple instances of GenomeBrowse and other Golden Helix products to download files from the Data Source Library simultaneously.
Console view polishes:
- Added a note for transcripts with missing start and stop codons.
- Now suppressing long transcript nucleotide sequences with an added link to expand when necessary.
- Fixed SSL handshake issue after logging out and trying to log back in.
- Various left align edge cases were fixed when converting data to TSF format.
- Exporting data in the VCF format from the data source library now:
- Correctly places the RS ID in the identifier field instead of creating a new INFO field.
- Removes white space from symbol names and replaces with underscores
- Removed preview boxes for the Expression Editor when adding a Filter by Sample in GenomeBrowse for Variant Maps and Heat Maps.
- Fixed left-align on creation of TSF files to handle more edge cases correctly.
- Added a “slice” function to the expression editor to strip out characters from data. This would allow values such as “[A/G]” to be treated like A/G for styling purposes.
- Feature List now remembers the previous maximum number of features displayed for the same source. Now add additional features in chunks of up to 10,000 features.
- Data Source Library export a VCF file in the VCF format now correctly copying format field information. Some numeric array fields were being classified as string arrays.
- Fixed font in data console for source information.
- Fixed Amino Acid rendering in MT Chromosome. Now, the amino acids in MT use the MT table instead of the standard amino acid table. See: https://www.mun.ca/biology/scarr/MGA2-03-28_mtDNA_code.jpg
- Convert Source Wizard bugs fixed:
- Allow segments/scaffolds/chromosomes to be skipped in the segment list.
- Fixed conversion for GTF files with no exon features to an annotation source this should enable the Convert Source Wizard to work for GenCode GTF files.
- Fixed detection of list/array fields if the first occurrence of a list is not in the first 1000 lines, also the converter now properly handles a field that contains a mixture of no values, a single value and lists of values.
- Suppress incorrect “GRCh_38” assembly from the assembly list if present.
- Prevent crash when sorting by a categorical array field in the Feature List.
- Fixed Export to VCF from an annotation source, prevented a crash when the doc string is missing.
- Fixed downloading annotation sources from the Data Source Library on Mac OS X.
- The shipped RefSeq Genes 105, NCBI source was indexed to allow searching on gene and transcript fields.
- Prevented searching non-indexed annotation sources to avoid cluttering the Search and Location Bar.
- Fixed crash when visualizing a region that included a coverage transform at the exact boundary of a chunk of data.
- Updated the default recent genome assembly list to include the three most recent human genome assemblies, and the most recent mouse and rat assemblies.
- New default RefSeq Gene Annotation Source based on GRCh37_g1k build curated directly from NCBI.
- Changed system default genome assembly to GRCh_g1k to take advantage of the best mitochondrial reference sequence and the new RefSeq gene annotation source with updated mitochondrial gene annotations.
- GenomeBrowse Source Convert Wizard
- can now handle enumerated or string lists for features.
- can compute extents for very small float64 values (i.e. p-values) on Linux/ Mac OSX.
- GenomeBrowse bugs fixed:
- Save as Image will now respect the current zoom if the region was obtained by entering the location in the genomic location bar without also scrolling to zoom.
- Deleting one or more plots by selecting them in the plot view and pressing the delete key no longer crashes the program.
- Fixed GTF source conversion to adjust CDS start/stop based on frame and allow for CDS to jump introns.
- Fixed crash when reordering plots on MacOSX by clicking and dragging one or more plots in the plot view and clicking on the GenomeBrowse window before the plots finished drawing.
- Added option for variant sources to left-align indels using a reference sequence source.
- Added the ability to save annotation sources as VCF, XLS, FASTA, and WIG files.
- Updated and expanded the functionality of the Expression Editor for filtering and added it to the Source Convert Wizard. Now fields can be computed based on existing fields for either filtering or adding to annotation sources.
- Fixed the following GTF source conversion problems:
- Exon starts and stops were listed in strand order instead of genomic order. This only resulted in issues when performing Variant Classification using these files.
- Accounted for out of frame transcripts to result in fewer invalid transcripts.
- Added source field GTF file source info.
- Fixed Convert GFF Files to Annotation Track to work on files that do not have mrna features.
- Replaced tabix with htslib to handle symbol collision. This should fix all crashes that were occurring when trying to compute coverage and index on BAM files and indexing and compressing VCF files.
- Fixed help links for data source library, source convert wizard and on plot help links.
- Fixed click issues for Marker Blocks on LD plots. It should now be easier to select the correct marker and marker block.
- GAserver has been updated to handle alias chromosome naming for remote sources.
- Added the “*.fna” extension to FASTA file options for the source convert wizard.
- Added Active/Passive FTP option for loading BAM files from URL.
- Delayed launching Evernote Note until after an authenticated connection is made. This fixes a problem when the note for a project failes to load on startup for some users.
- Fixed loss of dock size information on minimize/restore of GenomeBrowse main window.
- Fixed computation of BAM indexes and coverage files by upgrading to the latest htslib library.
- Fix regression where only one computation was run on a source such that a source requiring an index and coverage computation would not kick off the coverage after the index finished.
- Pileups now update after changing options without needing to force an update.
- Made it possible for interval tracks with zero width features to label all non-zero width features.
- Have assembly aliases used consistently in readers to re-map segment names.
- Made sure that each GenomeBrowse project remembers whether or not the feature list was shown or hidden.
- Fixed bug that prevented the tabix TBI file from being generated for a >2GB bgzip compressed VCF file on windows.
In version 2 we have rebuilt most of our backend and almost all of our frontend of GenomeBrowse with too many changes to count. Here are the large features you will notice that are new and changed.
- The “Add” sources dialog has been redesigned to easily allow managing multiple local and remote sources in one view. It features hierarchical repositories, a information pane for selected sources and the ability to plot other fields in a source other than the default (for example, plotting the allele frequencies from 1000 Genomes track instead of the variants).
- From the Add dialog, there is a new Convert Wizard to convert any tabular file as well as all our supported file formats to the new compressed and index TSF format. All public annotations are now also in TSF, which can be 1/10 the size of their previous versions. This includes the ability to import custom genomes from FASTA files and define a new genome assembly.
- The public data repository now is organized with folders to make it easier to find a specific type of data. Also, by default only the latest version of an annotation source is shown.
- Added ability to save plots as images. You can right click on individual plots or save all the plots in a project.
- Added documentation in the form of a GenomeBrowse manual as well as on plot help controls to jump to specific sections of the manual referring to the plot type.
- Added Evernote integration. You can create new notes or open existing ones from your notebook once you connect your Evernote account. The note editor is attached to the project and can be used to add bookmarks and insert images from your current context.
- Added IGV Session file support. All sources in the IGV session file that are supported by GenomeBrowse are added to the current project.
- Added support for streaming BAM from HTTP and FTP. Just click the URL button in the Add dialog and enter the URL of either a single source, or the index file or directory that contains BAM files. It’s required that each BAM has a corresponding ”.bai” file.
- Support for loading sources and changing the zoom of GenomeBrowse from other programs has been added in a number of forms. The installer now registers the genomebrowse: protocol and URLs can control a new or existing GenomeBrowse process. Also, under the Program tab of the Options dialog, you can enable a Link Server that by default listens on the same localhost port that IGV runs on, allowing programs that control IGV through http://localhost:60151/ urls to control GenomeBrowse.
- Numerous other optimizations, polishes and bug fixes.
- Added support for reading sorted BED files directly! The “track” line is respected to support display name, coloring and the bedDetail format variant.
- A new table view of a source’s features is available on the top-left of plots when you hover over them. You can view the features based on the current zoom or from the start of the source. Selecting features in this view jumps the zoom to that position.
- Added more details to the data console output for BAM coverage plots. Percentage of total reads is now an additional column for each nucleotide and insertion detected at the current position.
- Added an additional option to choose whether to automatically display of Coverage and/or Pile-up plots when adding new BAM sources. If either coverage or the pile-up plot is not displayed automatically, it can be shown by checking the appropriate box in the plot tree view.
- The “garead” and “gautil” command line utilities are bundled with GenomeBrowse. Running gautil precompute <source.bam> will produce index and coverage files. This may be useful for pipeline environments to prepare files for GenomeBrowse visualization.
- Some glitches with parsing the input to the location bar have been resolved. It has also been improved to be able to take whitespace delimited region descriptions like “chr7 123,456 444,555” as well as many others. As you type, the normalized range that was detected is shown in the results drop down so you know exactly what range will be used when you hit Enter.
- Overhauled the touch pad input handling on Mac. Zooming is significantly more responsive. Two finger scroll up and down zooms in and out while side to side pans the view.
- Before registering, you can now open a cloud-based demo project to give GenomeBrowse a test drive.
- Show transcript names as a hover-label when looking at whole genes.
- Fixed some crashes and unexpected behavior with the new downloader engine.
- Fixed issue where you could not log in if the “Stay Logged In” checkbox was not checked.
- Fix display of data for VCF coverage at chromosome level scale for sparsely populated sources.
- Reverse-strand hard-clipped reads were not being aligned correctly and thus looked like they had a lot of mismatches.
- Updated the links for gene names in the data console.
- Completely new downloader engine that supports download acceleration, pause/resume, auto-resume and can be minimized to the system tray for long downloads.
- Files are now downloaded with their precomputed coverage or indexes so they are immediately displayable.
- From pipeline.goldenhelix.com, ghdownload will now open the new GenomeBrowse downloader.
- Support for VCF Files! New rendering mode for multi-sample “Variant Maps”, as well as auto-compression/indexing of VCF to Bgzip/Tabix format required to read files directly for rendering.
- Single sample VCFs, multi-sample VCFs and even “site”-based VCFs with no sample data are now supported.
- SNV and InDels supported. We will do our best to draw Structural Variants, but their representation in VCF format is less standardized.
- GenomeBrowse should now start faster once you have downloaded the reference sequence track for your existing project.
- We have experimental support for “Value” tracks. More support coming in the near future.
- Many polishes and bug-fixes, including improved controls for changing labels on features.
- Gene tracks have been much enhanced. Exon numbers and transcript names are labeled at appropriate zoom levels as well as more details are provided while hovering over codons.
- The reference sequence track gives you forward, reverse and all theoretical amino acid encodings as you expand its height to give it more space to draw.
- Improved Y-axis labels, which is especially important when looking at VCF tracks with sample names as Y-axis labels.
- Added the “g1k” Human Reference build, which is GRCh37 with the rCRS MT as specified by the 1000 genomes project.
- Updated the genome assembly file format to handle more meta-data. Also improved when warnings are displayed for a source not matching the selected build.
- Added ability to change the field used to label features for Interval and Variant plots.
- Polished the label drawing system for all plot types.
- Added ability to filter low quality alignments from pileups and coverage plots.
- Polished rendering of BAM plots and mouse-hover capabilities.
- Performance improvements when rendering views with tens of thousands of reads.
- Significantly improved “gear” menu and control panel capabilities of plots.
- Fixed some crashes in pre-compute reported by users.
- Added an Update Available notification.
- Add ability to have project documentation.
- Have “demo” mode for putting on thumb drives.
- Added project management! Create new projects, save and open recent projects.
- Multi-window support: can open multiple projects in multiple windows.
- Added Illumina BaseSpace integration in the ‘Add’ dialog.
- Amino-acid information now drawn on genes. Also labeling of gene parts enhanced.
- Fixed crashes associated with attempting to index or run precompute on an unsorted BAM file.
- Fixed crashes when precomputing on certain BAM files.
- Prevent cache system by being overloaded by 30K+ read-depth regions. This should help with out-of-memory errors when viewing areas ultra-high-read-depth.
- Fixed issue where the directory to store settings and the default project was not created. You should now not need to log in every time you start GenomeBrowse and your project should save on close.
- The message boxes and progress dialogs have been updated to have a more consistent style.
- Made the Windows installer able to be run by a non-Administrator user.
- On Mac, use the native directory chooser so you can pick mounted volumes.
- Prompt the user to download a reference sequence track if there is not one locally when switching genomes or on first use.
- Fixed some crashes on Mac OS X when switching to and from GenomeBrowse.
- Fixed plots that got stuck in “Initializing...” mode.
- A few fixes of handling views of edge cases.
1.0.0 - Yea!¶
- Complete overhaul of the zoom system. Scroll wheel zooming on pile-up beautifully scales the content to an ideal aspect ratio.
- Login and registration window in place.
- Improved drag-n-drop of plots by their handles.
- Per-base quality information is used to shade mismatches in both pile-ups and coverage views.
- Many tweaks, polish and optimizations.
- Improved interactive mouse support for insertions. Labels displayed on hover for pileup and coverage plots.
- Polish and performance improvements to BAM rendering plots.
- While Y-axis zoom lock is enabled, smoothly adjust the y-limits as you pan into areas with high and low coverage.
- Significant performance improvement for Mac renderings, and improved handling of touchpad inputs like pinch-zoom and panning.
- Added interactive mouse support for the labeling system, allow you to hover over genes or variants and see their label even if it was hidden by default.
- Fixed crash when viewing certain regions of the 1000 Genomes track from our annotation server.
- Certain types of BAM files were creating invalid BAI files, which is now corrected.
- Fixed UI freeze that occurred when you tried to delete a plot that had a computation in the queue to be run.
- Many stylistic improvements and user interface polish and tweaks.
- Detect sample.bai as valid BAI file instead of just sample.bam.bai
- Larger zoom limits in pile-up mode before switching to drawing coverage
- Updated logo and name
- New <a href=”http://pipeline.goldenhelix.com/“>pipeline.goldenhelix.com</a> account integration. Redesigned the “Add” dialog to have tabs for Annotations, EA Pipeline Account and Example Samples.
- The Landmark view displays the Cytobands track by default, but can be set manually to show other plots.
- Totally new labeling system for genes and variant tracks. You’ll notice transcripts are grouped by their gene and have labels on the exons and utrs. Genes also collapse transcripts when vertical space is tight.
- The BAM pile-up plot now has a “gear” icon in the top left that allows you to switch coloring modes to emphasize mismatches versus strands. Also “Split by Strand” stacks forward and reverse strands above and below the X-axis.
- Tools->Proxy Settings allows you to auto-detect or manually enter proxy settings.
- Many stylistic improvements and user interface polish and tweaks.