Bug Fixes

• Added missing context help for changing group source in the Sequence Data Explorer
• Updates for the Sequence Data Explorer to make changing group source easier
• Fixes an invalid type cast bug in the Distance Matrix display when calculating averages
• The function for getting available memory on macOS has been updated to use the vm_stats command
• Fixes a potential range check error when setting the number of threads to use when doing bootstrap tree construction

Bug Fixes for macOS

• Added a workaround for tabs in the Tree Explorer not rendering correctly - using radio buttons instead for now
• Fix positioning of status bar in the Tree Explorer
• Fix sizing of first 6 columns in the Alignment Explorer for macOS

Bug Fixes

• Update group taxa count display in the Sequence Data Explorer when changing group source (continent, country, city, etc...)
• In the Tree Explorer, correctly toggle the checked state of divergence times and ancestral states actions when changing the tree display
• In the Sequence Data Explorer, refresh site coverage when user-specified site coverage percent is changed
• When doing RTDT (RelTime with Dated Tips) from the Tree Explorer, do not arrange taxa by input order as it breaks outgroup drawing
• For MUSCLE sequence alignment, do not replace missing/ambiguous base symbols (e.g. ?) with N or X (for nucleotide and protein data respectively)
• For MUSCLE sequence alignment, warn the user if stop codons are found and allow them to abort if desired
• Fix for parsing NEXUS files where the internal data structure used for Characters blocks might not be set leading to an access violation

Bug Fixes

• Fix drawing of outgroup in Tree Explorer when doing RTDT (RelTime with Dated Tips) analysis
• Add missing option to hide overlapping divergence times to Tree Explorer toolbar

Bug Fixes

• Fixes replacement of ambiguous sites for MUSCLE sequence alignment
• When doing MUSCLE alignment, check for stop codons and warn the user if they are found
• For parsimony trees, keep branch display action disabled until branch lengths are computed.
• For parsing NEXUS files, fixes a potential access violation when reading a characters block
• For spreadsheet-like exports, automatically updated file extension when changing export type
• When exporting alignments to NEXUS files, treat spaces in names uniformly across all blocks.
• Adds context-sensitive help for the CorrTest (rate correlation test) analysis
• Fix an access violation in the Calibration Editor when drawing update events fire before data is finished loading
• Updated citations for Corrtest and RelTime using calibration densities as those papers are now published in Molecular Biology and Evolution
• Fixes malformed example files in the NeiKumar2000 directory.
• Fixes alignment offset for concatenating files with different taxa names and addes checking that parsing of MEGA files succeeded.
• Fixes window freezing when using the ML framework and a user tree is provided automatically by the topology editor
• When scalebar font changes, set the position correctly
• Only allow timescale display for time trees.
• Fixes drawing the time scale for RTDT (Reltime with Dated Tips) trees
• Add missing help page for calibrating timetrees using a molecular clock.

Updates

• The RTDT (RelTime with Dated Tips) method for constructing timetrees calibrated using sample times from molecular data has been updated to support calendar dates in addition to sample years. Previously, tips could only be dated by year but now they can be dated by year, month, and day
• The Maximum Likelihood (ML) framework in MEGA has been optimized to use memory more efficiently so that larger data sets can be successfully analyzed in MEGA
• All distance calculations have been moved from the main GUI thread to separate threads so that the GUI remains responsive when working with larger data sets
• Additional meta-data commands have been added to the .meg sequence alignment and distance matrix file formats. These meta-data commands allow for dynamic grouping of taxa and are supported for group-wise analyses in MEGA
• In the Sequence Data Explorer, sites can now be automatically labelled based on site attributes (e.g. variable sites, parsimony informative sites, etc...)
• The Sequence Data Explorer has been updated with many minor enhancements such as navigation by highlighted sites, preservation of codons when exporting highlighted sites, highlighting labelled sites, selecting sequences based on group attributes, and other additions
• A feature has been added to the Tree Explorer for navigating ancestral state changes between nodes for Maximum Likelihood and Maximum Parsimony trees
• The Tree Explorer formatting toolbar has been completely revamped to be more user-friendly
• A feature has been added to the Tree Explorer to auto-collapse nodes by group name, by cluster size, or by branch length difference

Bug Fixes

• Fix to correct drawing of compressed clusters so that they are sized according to the taxa within the cluster
• Improved validation of calibration constraints in the Calibration Editor to more cleanly handle duplicate calibrations

Bug Fixes

• Fix for memory management when computing codon usage with an alignment that has gaps
• For exporting composition statistics results, let the user display results or save results to a file
• Updated the help for the command line version to describe RTDT (RelTime with Dated Tips) and added example files for RTDT
• Handle invalid timetree calibrations more cleanly by disabling calibrations that do not map to active taxa and add an action to remove invalid calibrations
• Fix progress dialog when computing alignment site attributes in the Sequence Data Explorer so that progress does not appear in front of other applications

Updates

• In the Sequence Data Explorer, users can select grouped sequences based on group size and largest N groups
• In the Sequence Data Explorer, users can now disable all sites without labels, effectively compressing data on-the-fly
• When checking for identical sequences when doing Reltime, check sequence entropy to minimize the number of actual string comparisons done as this is faster, especially when sequences are highly similar
• In the Sequence Data Explorer, added options to labels sites manually and to highlight labelled sites
• Optimized parsing of FASTA files for the Alignment Editor when working with very large data files. Several tasks, such as loading FASTA files and checking sequence names now happen in a thread outside of the main GUI thread
• Add number of selected taxa and groups to the status bar in the Sequence Data Explorer
• In the info window for the Tree Explorer, a list of all taxa under the selected node or branch is now displayed
• When the user searches for motifs in the Sequence Data Explorer, text displaying the number of occurrences found for a given motif is displayed
• RTDT times are now formatted as dates in the tabular export files, days elapsed have been added to the tabular export files, and tree session files have been updated for RTDT
• In the Tree Explorer info box, RTDT divergence times are now displayed as formatted dates instead of floating point values. Days elapsed since since the latest sampling time are also displayed
• Sample times for RTDT calibrations can now include the month and day in addition to the year for fine-grain calibration. Useful for fast evolving data

Bug Fixes

• Fixes a hanging thread (TUpdateHeaderStringThread) on Linux when using the Alignment Explorer
• For Reltime and RTDT, validate outgroup and ingroup before adding constraints when using the TimeTree Wizard
• Fixes rendering issues for Tree Explorer session files.
• Fixes display of outgroup when rooting trees in preparation for Reltime. Was showing multiple group brackets in some cases
• When searching for calendar dates while parsing tip dates from sequence names for RTDT, find overlapping regex matches
• Generate Reltime trees from within the Tree Explorer window using the same branch style in the Reltime tree that was used in the original tree
• In the Tree Explorer, added missing functionality to edit group and subtree formatting options
• In the sequence input data options, remember which data type the user previously selected
• In the Alignment Editor, when inserting sequences from files, display progress correctly
• In TAlignGrid, whenever showing a progress dialog, hide the cancel button because we have no handler for the button click
• For TAlignGrid, stop the TUpdateHeaderThread every time alignment data is modified, otherwise there can be range out of bounds errors
• In the alignment editor checking for delete gaps enabled now only happens on after selection instead of actionlist updates so the gui does not freeze for very large data sets
• Bug fix for sequence entropy calculation: check for zero base counts to avoid taking log of zero
• Fix list index out of bounds error for uninitialized status bar statistics
• In the Genes/Domains dialog on the Site Labels tab, only display the first sequence because for large alignments the draw grid is eating up memory
• When editing names in the Tree Explorer, set focus on the name edit box, otherwise, the toolbar gets key press messages
• For the site pick form only show one row of sequence data in order to conserve memory usage for large data
• In the Sequence Data Explorer, prevent row moving when user is just trying to check/uncheck sequences
• For doing Reltime OLS, the calculation for the distance matrix now executes in the Reltime search thread instead of the GUI thread
• Checking for identical sequences for Reltime has been sped up and progress is now shown when loading and validating calibration constraints
• For ClustalW alignment, don't disable Alignment Explorer because the progress window can get trapped behind after switching between applications
• Fix an access violation in Sequence Data Explorer after exporting data
• Fix access violation in Tree Explorer where extended character format for ancestral seqs not disabled for ML
• Fixed a bug in the Reltime analysis where an ancestral node could have a time smaller than a descendant node
• Fixed incorrect MP search level and missing initial tree information for the caption for ancestral sequences
• Fix a bug where when saving a session with translated data from the SDE, when loading the session, translation was not done leading to index out of bounds errors
• Perform text search in the web browser by using Javascript in the TFindDialog.OnFind event because the Chromium text search was not working
• Updates the sequence name option dialog to correctly use the option to abbreviate the genus name and also updates the drop downs to update the proposed sequence label when new selections are made

Updates

• Added support for rooting user input trees using an outgroup for ancestral sequence inference
• For ML phylogeny inference, use the selected model for auto-generating the initial tree if the model is available for distance methods
• For the Reltime analysis, fine-tuning on confidence interval correction for shallow nodes has been implemented

Bug Fixes

• Fixed export bug in the Sequence Data Explorer where the export thread could hang
• Fix for alignment session files (.mas and .masx files) where MEGA could read past the end of the file
• Fix for exporting SVG images from the Tree Explorer where existing SVG files were not shown in the save dialog
• Disable the site number spin edit in the Tree Explorer when there is only a single site
• For user-provided input trees, do not root on the midpoint when displaying in the Tree Explorer
• Do not prompt the user to close active data when opening a Newick file by double-clicking on the file
• Turned of branch length estimation in MEGA-CC when doing Maximum Parsimony tree construction
• Fix to not update the Alignment Explorer header string until after actions are completed so that Index Out of Bounds errors are avoided
• Fix for pasting sequence data in the Alignment Explorer
• Fix for exporting trees to SVG files where the command to draw the full tree was not set
• Fix for tree session files (.mts and .mtsx) where a change in the size of enumerated types led to failure to read these session files
• When opening an alignment session file for analysis rather than for alignment, use the correct parser
• Fix for Clustal coding alignment where the Alignment Editor could not receiv focus
• Fixed range check error on Linux when getting maximum heap usage
• Do not use rounding for Reltime calibration density samples

Updates

• Added help documentation for the alignment concatenation feature

Bug Fixes

• Fixed a bug in the ML framework where on rare occasions, branches near the base of the tree were collapsed
• For the macOS version fixed a bug where in the Alignment Explorer pasting data copied from external applications did not work
• Fixed a crash bug when exporting PDFs from the Tree Explorer when using non-standard fonts
• Fixed a floating point overflow bug for ML log likelihood
• Handle errors gracefully when users try to use the macOS version of MEGA to open tree session files made on Windows systems
• Added validation of pen, brush, and font styles when loading tree session files to catch errors when loading tree session files across CPU architectures (e.g. session files made on 32-bit systems should not be opened on 64-bit systems)
• When showing error messages from tree search threads, switch to the context of the main thread to avoid crashing
• Fixed a crash bug when exporting newick files after doing RTDT using the OLS approach
• Added missing progress updates when generating a global clock tree
• Added missing prompt for outgroup when doing ML user tree analysis with option to assume a global clock
• Fixed a bug where the degrees of freedom for the molecular clock test was incorrectly reported
• Fixed a bug where cancelling the outgroup prompt for the ML Reltime analysis did not correctly update the number of currently running ML analyses
• Updated parsing of Genbank data to handle missing expected fields

Updates

• Added an option to use the log-normal distribution for calibration densities with the Reltime analysis
• Added a check to warn users when identical sequences (after data subsetting) are found when using the Reltime analysis as identical sequences are problematic for the Reltime method
• Optimized cell drawing in the Alignment Explorer to better support large data sets

Bug Fixes

• Changed the labels for visual controls in the Calibration Density window to more cleary indicate that a statistical distribution is used to generate calibration densities
• Fixed a bug where using the integrated web browser to import Genbank formatted data with mix of sequences with and without CDSs skipped the sequences without CDSs
• Refactored the import of sequence data from the integrated web browser to be asynchronous because macOS systems will not allow modal dialogs in the context of the browser process
• Fixed mouse down event handlers in the Alignment Explorer for macOS which had broken sequence name editing as well as copy/paste operations
• Fixed a bug in MEGA-CC for getting the number of threads to use for the Find Best DNA/Protein Models (ML) analysis
• Catch division by zero conditions for the Estimate Position-by-Position Rates (ML) analysis
• Bug fixes for using the Parse Tip Dates dialog for the RTDT analysis when running on Linux systems
• Fixed the size and positioning of controls in the Analysis Preferences Dialog for Parsimony tree construction when in Prototyper mode
• Fixed the caption for the Test Molecular Clock (ML) analysis where the number of parameters for the tree with clock was incorrect
• Fixed a memory leak in the Construct/Test Maximum Likelihood Tree analysis
• Fixed the rendering of menu items so that accelerator keys are drawn correctly
• Fix for setting the number of decimals to use for displaying branch lengths in the Tree Explorer window
• Enabled the cancel button in the Tree Editor form
• Bug fix for the Trace Edit Form not opening on macOS systems
• Added checks for floating point underflow/overflow for Maximum Likelihood analyses

Updates

• The installer package and executables have been notarized to meet the elevated security requirements introduced in macOS 10.15 (Catalina)

Bug Fixes

• Fixed a bug where the Trace Editor did not display properly on macOS systems
• Fix for Tree Edit Form where clicking the cancel button did nothing

Updates

• The integrated web browser was updated to work with the new NCBI BLAST results page for importing sequence data
• The optimization of branch lengths when using the Maximum Likelihood method has been improved
• Updated the spreadsheet writer to support spreadsheets with 16,384 columns instead of just 256

Bug Fixes

• Fixed drawing of bootstrap values in the Tree Explorer to locate the values more accurately
• Fixed drawing of bootstrap values in the Tree Explorer where text was not rotated for some fonts
• Fix to handle underscore characters in taxa names for time calibrations when doing the RTDT analysis
• Updated the file iterator system in MEGA-CC to try and continue over all input files specified even if some fail to be processed
• Bug fix for exporting pairwise times when a newick file was loaded instead of a constructed tree
• Fix for tree session files (.mtsx) to remember the previously used font for taxa names
• Fixed a bug in the ML framework that could cause likelihood values to become zero for problematic data sets

Updates

• The macOS version is now 64-bit and runs natively using the Cocoa widgetset. WINE and Wineskin are no longer used
• Added the RTDT (RelTime with Dated Tips) method for building timetrees using tip date calibrations (see Miura et al. 2019)
• Added support to RelTime for using calibration densities
• Added support for calibrating RelTime trees using a global evolutionary rate or single fixed divergence time
• Added a utility to infer tip dates from sequence names
• A utility to export pairwise distances for any tree with branch lengths has been added to the Tree Explorer
• For the Windows version, export to the EMF (Enhanced Meta File) graphics format has been added to the Tree Explorer
• Improved support for high resolution monitors
• MEGA CC for Linux can now run on headless machines (removed dependencies on GTK)
• Added support in MEGA CC for site labels
• The macOS version now uses native keyboard shortcut commands

Bug Fixes

• Fixed List Index Out of Bounds error for NJ tree construction on Linux
• Fixed multiple column selection in the Alignment Explorer
• Fixed otu name font not used from saved tree session file
• Fixed freeze in Tree Explorer for very large trees
• For MUSCLE sequence alignment added support for non-Latin characters in path names
• Added window session properties to the explorer windows to remember size, position, font, etc...
• Removed annoying popup window when creating a new tab in the integrated web browser
• Corrected ML variance estimation for nodes with zero data coverage
• Improved rendering in Tree Explorer for macOS systems
• Fixed access violation when running MEGA-CC in silent mode
• For the macOS version, fixed an issue where overlapping divergence times were not hidden
• Fixed copy, cut, and paste issues in the Alignment Explorer
• For ML ancestral sequence inferrence, exclude the root node

Updates

• Refactored the PDF export from the Tree Explorer to produce a true vector graphics file
• Added SVG (Scalable Vector Graphics) export of tree images from the Tree Explorer
• Implemented text search in the Mutation Explorer
• Added copy-to-clipboard for the Mutation Explorer
• Added spreadsheet export for results from several analyses (Reltime analysis, molecular clock test, Tajima neutrality test, Tajima clock test)
• Added the alignment concatenation utility to MEGA-CC

Bug Fixes

• Force the use of the . character as the decimal separator to resolve localization issues
• Improved the format of data copied to the clipboard for results captions
• Added missing menu click events for sorting in the Sequence Data Explorer
• Improved the positioning of subtree captions and images in the Tree Explorer
• Bug fix where group names consisting of a single character broke the parser for .meg data files
• Bug fix for OLS user tree analysis to correctly handle the case where a user tree with no branch lengths is provided
• Bug fix for drawing circular trees when showing the topology only (not scaled) - branches were disconnected
• Fixed integer overflow error in the Tajima neutrality test when using very large input data
• Moved Tajima clock test to its own thread so that the GUI is responsive when using very large input data
• Fixed drawing of the focus in circle and curved line trees
• When drawing caption brackets in the Tree Explorer, account for node IDs if they are displayed so that brackets are positioned correctly
• Implemented missing copy-to-clipboard utility for the Trace Edit form

Updates

• Added Evolutionary Probabilities calculation (see Patel and Kumar 2019)
• Added Codon by Codon Selection analysis using the SLAC analysis at www.datamonkey.org
• Updated the Analysis Preferences Dialog to initialize with smarter default options

Bug Fixes

• Fixed linearization of UPGMA trees in Tree Explorer
• Fixed MUSCLE codon alignment bug
• Re-factored progress reporting for Model Selection analysis to resolve potential deadlock
• Fixed index out of bounds bug for Fisher's Exact Test
• Improved positioning of Ancestral state characters in the Tree Explorer
• For .deb packages, add libgconf-2-4 to dependencies as it is needed on some Ubuntu systems
• Updated MUSCLE options to use correct default values for open gap penalty

Updates

• Added Corrtest (rate correlation test) analysis (see Tao et al. 2018)
• In the Alignment Explorer, scrolling is now much smoother
• Added a notification to the main form for MEGA-CC versions so the Prototype and Analyze modes are less confusing
• Added coding DNA option back to Prototype mode

Bug Fixes

• Fixed ClustalW alignment bug where alignment grid would sometimes add an extra column
• Updated ClustalW options to add missing default option values
• Fixed sporadic 'session file not found' message when opening Newick tree files
• In Alignment Explorer, hide the groups column if no groups are defined
• Fixed copy to clipboard bug in the Trace Editor window
• Bug fix for Reltime - do not allow a max time constraint at the in-group root
• Fix for printing captions from the Tree Explorer

Updates

• Re-factored alignment export from the Sequence Data Explorer to handle very large files
• For MEGA-CC users, a standalone binary executable that does not require installation is now available for download

Bug Fixes

• Fixed printer setup error
• Added missing Print function in the Trace Editor
• Fixed bug where GUI was unresponsive when adding protein data to nucleotide alignments
• Fixed printing trees from the Tree Explorer
• Copy to clipboard in Tree Explorer now works
• Fixed MUSCLE alignment crash on Ubuntu
• Added missing man pages for Linux builds
• Disable ancestral states function in Tree Explorer when that information is not available
• Don't allow multiple instances of MEGA X to run

Updates

• Added a Prototype mode so that MEGA Analysis Options (.mao) files for MEGA-CC can be generated in MEGA X (this replaces MEGA-Proto which will no longer be supported)
• MEGA-CC is now distributed with MEGA so that MEGA can be used as a full GUI application or a command-line application

Bug Fixes

• Numerous bugs in MEGA X have been fixed. Many thanks to all who helped test the development version of MEGA X!

Updates

• MEGA GUI now runs natively on Linux in addition to MS Windows
• Bootstrap tree construction when using Distance and Parsimony methods now uses course-grain parallelization (multiple bootstrap replicates run simultaneously) to optimize performance
• The model selection analysis now uses course-grain parallelization (multiple models are tested simultaneously) to optimize performance
• The user interface (UI) has been modernized
• A feature to concatenate sequence alignments has been added
• Added support for exporting spreadsheets to the Open/Libre Office (.ods) format
• Added support for exporting spreadsheets to the Excel XML (.xlsx) format

Bug Fixes

• Fixed a bug where the option to hide overlapping taxa in the Tree Explorer was not respected

Bug Fixes

• Fixed invalid floating point operation error in Reltime analysis
• Reduced pixel format for exporting tree images to PNG and TIFF files for larger trees
• Updated PDF export of tree images to auto-scale dimensions according to users dpi settings

Updates

• Changed file extensions for alignment and tree sessions to .masx and .mtsx respectively because Microsoft Windows is now associating the original file extensions (.mas and .mts) with media file formats
• Allow users to continue user tree analyses from the Topology Editor without forcing all sequences to be used or all tree taxa to be resolved
• Allow users to cancel long-running SPR and TBR parsimony bootstrap tree construction part of the way through and use only the reps completed as long as the min number of reps (50) have finished
• Added divergence time confidence intervals when Reltime is performed on trees generated using distance or parsimony methods

Bug Fixes

• Fixed memory leak when doing ML tree construction with bootstrap
• Bug fix for importing outgroup taxa names in TaxaGroupsDlg when doing the Reltime analysis
• Fixed potential division by zero error in Reltime
• Fixed bug where exporting captions as text would produce an empty result
• For ML analyses, validate that sufficient data is available for executing the calculation

Updates

• Added Nexus export of Reltime trees
• Optimized the Reltime variance estimation
• The Windows installer for MEGA-GUI now has an option to remove older versions
• Added progress updates for saving alignment sessions
• Updated the Tree Explorer to support taxa name editing

Bug Fixes

• Fixed a bug in the Applinker where the MUSCLE process would not terminate properly if the user aborted the alignment
• Handle the case where a user closes the active data file while phylogeny construction is still running
• Bug fix for ML tree construction where if the user cancelled the analysis, the calculation still went to completion
• Bug fix for the Alignment Editor where if trying to remove gaps fails, the editor was not re-enabled
• Fixes a list index out of bounds error when doing alignment with a blank sequence
• Do not allow users to open or add new data in Alignment Explorer while a sequence alignment is in progress
• Improved the warning message that is displayed when MUSCLE removes sequences - now only shows 1 message instead of multiple

Updates

• Reltime CI estimation has been improved for the deepest node
• Reltime calibrations are now saved in tree session files

Updates

• After running the Reltime analysis, it is now possible to recalibrate the timetree without running the Reltime calcuation anew

Bug Fixes

• Updated Alignment Explorer so that launching a BLAST search on NCBI correctly populates the input form
• Added a 'shutdown gracefully' thread that waits for any CEF browser threads to terminate before the main form closes to avoid potentially orphaned CEF browser threads

Bug Fixes

• Fixed one-off error in Alignment Explorer when translating selected sites

Updates

• Added drag/drop support to the text editor
• Added a button to the text editor to make closing individual tabs more intuitive

Bug Fixes

• Improved the calculation of standard error for divergence times for the Reltime method
• Fixed a bug in Analyze User Tree by Maximum Likelihood where the calculated log likelihood value could be incorrect for some datasets (although branch lengths are still correct)
• Disallowed opening of tree session files made in MEGA6.06
• Fixed a bug in the Taxa/Groups window where the icons for disabled buttons were not visible
• Added auto-pruning of any unused taxa from the user tree for the Reltime method
• Updates to the help documentation
• Fixed one-off errors in exports of ancestral sequences to Excel and CSV files

New Features

• MEGA7 is now optimized for 64-bit processors and can utilize a drastically larger amount of RAM
• The Tree Explorer has been refactored so that it can now display trees with > 100k taxa (dependent on the graphics capabilities and memory available on the host computer)
• Added a Timetree Wizard that simplifies the creation of timetrees
• Added support for rooting a timetree on an outgroup
• The timetree system now computes data coverage at all internal nodes in a timetree
• Added additional image export formats (including the return of the TIFF format)
• Removed the dependency on Image Magick
• Updated the Taxa/Groups dialog to support the definition of outgroups
• Updgraded the integrated browser from version 1 of the Chromium Embedded Framework to version 3

Important Notes

• MEGA7 will not open session files created with earlier versions of MEGA. Although inconvenient, this limitation is necessary in order to compile MEGA with 64-bit optimization. To retrieve data from session files made with previous versions of MEGA, open the session files with the same version of MEGA that created the session and export the data to .meg or .nwk files as needed. Then the data can be used in MEGA7.
• MEGA7 (and future versions of MEGA) is compiled for use only on 64-bit computers. Users that wish to run MEGA on 32-bit systems must use MEGA6 or earlier

Bug Fixes

• Fixed a bug in the text editor where the horizontal scrollbar would only scroll to 1024 characters
• Fixed a bug where unrooted trees exported to a Newick file would be rooted
• Moved the Examples folder from the Program Files folder (which can lead to access control issues) to the user's Documents directory
• Fixed one-off errors in the Sequence Data Explorer and Alignment Explorer where the last site in an alignment was sometimes not displayed
• Fixed typecast error when exporting results from the Caption Expert to Excel or csv files

Bug Fixes

• Fixed a bug in Alignment Explorer that would cause an access violation when editing the name of a blank sequence
• Updated the progress dialog for bootstrap to remove flickering and improve efficiency
• Fixed a bug in Distance Matrix Explorer where export of group calculations would fail if distance data was used
• Added download progress for the update system
• Fix for estimation of parsimony branch lengths when ambiguous data is present
• Updated the integrated browser to remove the cache folder that gets created on start up
• Removed the email option for sending bug reports (the email account is disabled due to spam)

Bug Fixes

• Updated the session system to open .mas files (alignment sessions) made in previous versions of MEGA (retrieves sequence data only)
• Updated the session system to open .msdx files (data sessions) made in previous versions of MEGA
• Fixed export of distance matrices for distances computed between groups from the Distance Data Explorer window
• Turned off bird's eye view in Sequence Data Explorer for short alignments (was causing rendering issues)

Bug Fixes

• Fixed a bug in Alignment Explorer where translating then un-translating a segment of data would lose data
• Fixed a bug in Alignment Explorer where the last site of a sequence would be displayed as a gap character
• Updated Alignment Explorer to use darker text for highlighted cells making them readable on Windows 7+
• Fixed one-off error for cut/copy operations in the Alignment Explorer
• Added missing resource files

New Features

• Added an option to open links in new tabs or new windows when using the integrated browser

Bug Fixes

• Fixed rendering of node height error bars
• Fixed a bug where a bootstrap value would disappear when re-rooting a tree
• Fixed memory leak during likelihood calculation

Bug Fixes

• Fixed a division by zero error for timetree analysis which would produce a message saying "NAN is not a valid floating point number"

Bug Fixes

• Fixed the web browser to open new pages in a new tab instead of a new window when 'target="_blank"'
• Fixed a bug where drag/drop of fasta files on the main form was not working
• Fixed caption error where if molecular clock was assumed for ML user tree analysis, the assumption of molecular clock was not mentioned
• Updated the installer to correctly remove all auxiliary files on un-install

Bug Fixes

• The installer in MEGA 6.0 Beta 1 didn't register dlls for image magick.

New Features

• MEGA6 contains a number of enhancements over MEGA5. A brief communication (MEGA6: Molecular Evolutionary Genetics Analysis version 6.0) describing the new features in MEGA6 has been published in the journal Molecular Biology and Evolution. A PDF of this brief communication can be downloaded here .

• Added a Timetree system for estimating relative and absolute times of divergence for all branching points in a phylogeny that is based on the Reltime method described in Tamura et al. 2012 and which supports multiple user-defined calibration constraints.
• Added support for internal node labels in Newick files in order to facilitate mapping of divergence times in MEGA's tabular timetree output. Internal node labels should be placed in single quotes.
• MEGA's capacity to allocate system memory has been doubled for 64-bit computers by setting the IMAGE_FILE_LARGE_ADDRESS_AWARE bit.
• The Tree Explorer was enhanced to allow for the display of trees with up to ~4,000 taxa.
• A bird's eye summary view was added to the Sequence Data Explorer. This view facilitates fast navigation across genes and domains.
• The LG amino acid model (Le and Gascuel 2008) was added for amino acid and protein-coding data.
• For Maximum Parsimony phylogeny construction, the CNI search method was replaced by the Subtree-Pruning-Regrafting (SPR) and the Tree-Bisection-Reconnection (TBR) approaches.
• Support for groups was added to the Alignment Explorer.
• Added support for exporting to fasta files from the Sequence Data Explorer.
• Added a utility for displaying node ids in the Tree Explorer (click View | Show/Hide | Node Ids)

Bug Fixes

• Fixed a bug where parsing of unaligned Fasta files sometimes failed.
• Updated HyPhy executable to the version 2.1.2. which includes numerous bug fixes, HBL feature adjustments, and stability improvements

Bug Fixes

• Fixed caption bug where caption occasionally showed up completely blank.
• Updated web-browser

Bug Fixes

• OLE Error in MUSCLE during alignment for MAC Fixed.
• Updated HyPhy executable to the version 2.1.2. which includes numerous bug fixes, HBL feature adjustments, and stability improvements

Bug Fixes

• Fixed bug in HyPhy related calculations persuant to gap handling.

Bug Fixes

• Fixed a bug in the installer that allowed multiple instances of MEGA.
• Updated the installer so it will correctly remove previously installed beta versions.

New Features

• Sped up ML bootstrap tree construction (by avoiding likelihood computation at non-resampled sites).
• Added TBR (Tree Bisection Reconnection) and SPR (Subtree Pruning Regrafting) search heuristics for MP phylogeny reconstruction.
• Removed CNI (Close Neighbor Interchange) search heuristic from MP phylogeny reconstruction.
• Replaced slow TIFF image export with PNG export (similar format, lossless, widely used).

Bug Fixes

• Fixed a critical calculation bug which occurred in some cases where bootstrapping wasn't resampled correctly (only was an issue with ML trees). This bug only ever existed in the beta.

New Features

• Enhanced Maximum Likelihood tree searching to utilize multiple processors/processing cores.
• Added Subtree-Pruning-Regrafting (SPR) search heuristic for Maximum Likelihood phylogeny inference.
• Expanded options for automatic initial tree generation for Maximum Likelihood phylogeny inference to include Maximum Parsimony.
• Added 'Branch Swap Filter' option to ML analyses which gives the user some control over optimization stringency with respect to branch lengths and likelihood improvements.
• Replaced IE based browser with a Chromium based browser.

Bug Fixes

• Bug fix for Excel export of results from HyPhy analysis where p-values were reported for codons that were not applicable.
• Fixed a division by zero error for some datasets when performing Maximum Likelihood 'Test Molecular Clock (ML)'.
• Fixed a bug where caption expert displays 'Data Specific model' instead of 'General Time Reversible model' for ML inference using GTR.
• Fixed the Analysis-->User Tree Computation-->Edit/DrawTree (Manually) menu item which was not linked to an event handler.
• Fixed the Analysis-->User Tree Computation-->Display Newick Tree menu item which also was not linked to an event handler.
• Fixed an access violation that occurs when displaying CaptionExpert for analysis of a user tree where assume molecular clock == true.
• Fixed a caption expert bug for Gamma Paramater estimation where the caption incorrectly indicates the source of starting tree.
• Fixed a bug that occurs when searching for a taxon in Tree Explorer so that MEGA will now jump to the taxon if found and if it is not visible in the current window.
• Fixed the search function in Tree Explorer so that all matches of the search query will be highlighted.
• Added a more graceful exception handling for malformed phylip sequence files.
• Fixed a bug that prevented opening valid phylip sequence files using the Alignment Editor.
• Fixed a bug where previously, if MEGA couldn't read a MEG file for reasons such as a locked file or one located on a network drive MEGA would crash instead of notifying the user of the problem.
• Fixed a bug in the App Options Dialog which wasn't correctly enforcing minimum and maximum constraints for values that are keyed in (rather than using the spinners).
• Fixed bug where site labels were 1 column off in Tree Explorer.
• For the App Options Dialog used with MUSCLE, removed the confusing checkboxes and updated the default values to reflect what they should be, as recommended by MUSCLE's author.
• For the App Options Dialog used with MUSCLE, added a 'Restore Defaults' button to restore the default values.
• Fixed MUSCLE alignment issue by accounting for sequences which may re-order themselves after alignment. This bug would result in sequence data with the wrong name attached to it.
• Fixed a bug with MUSCLE alignment where the removal of sequences using Alignment Explorer were not handled properly.
• Fixed ML export of most probable ancestral sequences where indels were not represented correctly.
• Fixed ML/MP export of ancestral sequence changes list where indels were not represented correctly.
• Fixed crash bug that would occur when trying to display the caption for NJ trees generated from distance data.
• Fixed export of caption for model selection where table values had no whitespace between them.
• Fixed caption for ME phylogeny inference where the caption always reported the search level used as zero.
• Fixed out of memory crash bug for ML and MP inference. Now if we run out of memory, the user will be notified the analysis is not possible.
• Fixed sporadic floating point overflow error in ML phylogeny inference.
• Fixed crash bug that would occur when closing the text editor after converting very large sequence data files to the Mega format.
• Fixed overflow that would occur when displaying trees with very long taxa names.
• Fixed overflow error caused by alignments whose length is less than the number of sequences in the alignment.
• Fixed the calculation of number of degrees of freedom used for molecular clock test.
• Fixed a bug where dialogs could become hidded behind various explorer windows if the user switches between applications.
• Fixed rare 'GammaQ' crash bug that could occur when doing linearization.

Bug Fixes

• Added additional information for automatic update system and fixed small bug.
• Simplified MUSCLE interface (removed checkboxes), updated default values based on MSUCLEs suggestoins.
• Fixed MUSCLE alignment re-ordering bug which was introduced in 5.04 which lead to sequences being associated with the wrong sequence name.
• Updated citations for Caption Expert.

Bug Fixes

• Added additional information for automatic update system and fixed small bug.
• Simplified MUSCLE interface (removed checkboxes), updated default values based on MSUCLEs suggestoins.
• Fixed MUSCLE alignment re-ordering bug which was introduced in 5.04 which lead to sequences being associated with the wrong sequence name.
• Updated citations for Caption Expert.

New Features

• Added option to turn off hiding of overlapping taxa. This is available from the View menu of the Tree Explorer.

Bug Fixes

• Fixed 'Export Most Probable' option where an error happened if you had selected all ancestors rather than 'Show most probable'.
• Fixed functions which grabbed sequences via the 'add to alignment' function.
• Fixed issue with the Tree Builder/Editor dialog where the scrollbar would be very wide, obscuring part of the tree.
• Removed -stable flag when using Muscle to align sequences upon the author's notification that the flag can cause an unknown bug, refactored MEGA to handle reordering of sequences.
• Fixed 2 problems in the .mts file. One in the generation of concensus trees, and another in the tree display.
• Added a progress dialog while downloading updates.
• Fixed export status message when exporting CSV, text, or MEGA format in the display matrix window.
• Fixed disappearing output type combo box in display matrix window, and added additional warnings if data is too large.
• Fixed Composition Distance analysis (previously displayed all 0s).
• Fixed Disparity Index analysis (previously displayed all 0s).
• Fixed Disparity Index Test analysis (previously displayed all 0s).

Bug Fixes

• Fixed bug which was accidently introduced only in 5.02 that effected linearization of trees.
• Fixed issue where some fringe datasets would cause slight linearization score drifts.
• Fixed some sequence importing from web browser.
• Introduced new MEGA for the mac system with improved reliability and addressed mac install issues.
• Released a Linux version for the first time since MEGA 4.0.

Bug Fixes

• Fixed tie tree problem.
• Fix for bug #0002672 Subject: Overwriting of sequences with duplicate names in fasta input.
• Fix for the number of parameters reported for JC in Best DNA/Protein Models

Bug Fixes

• Improved the initial tree making method for ML (now chooses between NJ and BioNJ trees based on superior likelihood)
• User Tree branch length corrected
• Bug fix to ignore user tree branch lengths
• Display fix for HyPhy output