which I assume has to do with how diamond finishes the daa file.
Error is hard to replicate because if daa-meganizer is executed again, no error is prompted and file meganizes normally. When visualizing in MEGAN, reads are overwhelmingly binned in high taxonomic nodes.
I’ve created a small database file and a small query if anybody wants to reproduce the error. Curiously, if you run daa-meganizer a second time on the failed file, no error happens and the file can be opened in MEGAN.
this does not appear to be a Diamond problem, this looks like a meganizer problem… meganizer writes reference annotations to the DAA file and then reads them in again to do binning and that is when the problem occurs.
I have just spent a couple of hours trying to fix it, without luck… Did you only observe this on this one file?