MEGAN5 potential misannotation


MEGAN5 is annotating the sequences in the attached xml file as an Arthropod taxa ‘Ala’ when in fact they should be placed within a bacterial category. I think this is because many of the hits are to:

“Uncultured SAR11 cluster alpha proteobacterium clone RS_10M_S22_61 16S ribosomal RNA, partial sequence; tRNA-Ile and tRNA-Ala genes, complete sequence; and 16S-23S intergenic spacer, partial sequence”

MEGAN5 may be finding the ‘Ala’ term and using that to assign the sequence? I’ve also attached the exported MEGAN taxonomy.

MEGAN command run over the command line:
import blastfile=‘Ala_18S_M2W_NT.xml’ meganFile=‘Ala_18S_M2W_NT.rma’ minScore=100 maxExpected=1e-25 topPercent=2 minSupportPercent=0 minSupport=1 minComplexity=0 lcapercent=85;

Thank you,

meganout_Genus.csv (264 Bytes)

Ala_18S_M2W_NT.xml (1.4 MB)

Dear Katie,

thanks for the bug report. MEGAN provides a mechanism for disabling taxa, as shown here.
First search for the label that causes a problem using an empty and using the uncollapse feature of the find dialog. Then use the Taxon Disabling menu.
Best wishes