BEAUti keeps old dates and locations when replacing aln

[carlopacioni]

When replacing an alignment after having populated the tip dates (and the location when I use the BDMM template), BEAUti generates a faulty xml file because dates (and locations) remains from the old aln even when I clear and regenerate them.

[rbouckaert]

@carlopacioni when using the Standard template, importing examples/nexus/Flu.nex, setting tip dates, deleting the Flu partition it is still possible to import an alignment without error messages, and produce a viable XML file. Is this a problem with BDMM template only perhaps?

[rbouckaert]

Another possibility is that the jblas library you mentioned (on the user list) is the cause of this problem and that this interferes with the normal behaviour of BEAUti. BDMM depends on MultiTypeTree, which uses jblas, so it may be possible that using the BDMM template triggers the jblas error.

[carlopacioni]

Hey remco, you are right when you remove the aln and then import a new alignment again the dates and possibly locations as well get parsed correctly but if you replace the aln clicking on the little r on the bottom left corner of the windows in beauty, the aln gets replaced but then I can't update the dates. that happens when I save the XML file before replacing the aln or when I load an XML file already been created earlier. I hope this helps. Carlo

On Tue, 25 Sep 2018 13:01 Remco Bouckaert, [email protected] wrote:

@carlopacioni https://github.com/carlopacioni when using the Standard template, importing examples/nexus/Flu.nex, setting tip dates, deleting the Flu partition it is still possible to import an alignment without error messages, and produce a viable XML file. Is this a problem with BDMM template only perhaps?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/CompEvol/beast2/issues/812#issuecomment-424191983, or mute the thread https://github.com/notifications/unsubscribe-auth/AGfu0e8EXheFZMbX15JfS9xpU4iqEe9qks5ueZyAgaJpZM4W18PL .

[rbouckaert]

Hi Carlo, Thanks for clarifying. You are right that the dates and other traits remain, and BEAUti should indeed warn and remove this info if the replacement alignment has taxa that were not in the original analysis.

[carlopacioni]

;-)

On Wed, Sep 26, 2018 at 8:07 AM Remco Bouckaert [email protected] wrote:

Hi Carlo, Thanks for clarifying. You are right that the dates and other traits remain, and BEAUti should indeed warn and remove this info if the replacement alignment has taxa that were not in the original analysis.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/CompEvol/beast2/issues/812#issuecomment-424517988, or mute the thread https://github.com/notifications/unsubscribe-auth/AGfu0fAhPBhN45_7wqNi_6stRljdVzCGks5ueqkLgaJpZM4W18PL .