Hi,
When I use merge.otus, I was able to merge for taxlevel 2 (phylum), 3(class), 4(order), and 6(genus), with taxonomy and shared files as input. But taxlevel 5(family) gave me the same result as taxlevel 6 in the taxonomy output file; the shared output file was different though. Is it a bug?
Best,
Yan
Hi - is it possible that your sequence data didn’t have different classifications for the family and genus level? For instance, if everything is unclassified at the genus level it would give the same results at levels 5 and 6. If you have examples of things getting merged at the family level that you don’t think should be getting merged, can you post them here?
Thanks,
Pat
Hi,
They can be classified at the genus level. When I did merge.otus for the family level. The taxonomy output that was supposed to at the family level was actually printed at the genus level. But the OUTs seemed to be merged for the family level, because less number of OTUs were printed for the family level merge than the genus level merge. Also, not all the genus levels were printed in the family merge output. So I guess OTUs were merged correctly at the family level, but it’s just the taxonomy levels were not correctly printed?
The first photo is for the genus level merge, and the second is for the family level merge. I combined the taxonomy and shared results into one excel file. As you can see the taxonomy ouput for the family level merge is at the genus level, but not all genes (e.g. red ones) were included.
Thanks -
I’m still not entirely sure what you’re seeing as the problem.
If I look at OTU16, which is a Commonadaceae, in the genus version it has a total abundance of 2526 and in the family level it has a total abundance of 11255 indicating that OTUs are getting merged at the family level.
Perhaps you’re seeing the genus names on the taxonomy from the family level merge and thinking that is a problem. For the data from the family level merge, you’ll need to trim back the string to the family name. If OTUs get merged then you won’t see the duplicate OTUs, only one OTU per family.
Feel free to clarify if you have examples of where merging is not happening.
Pat
Hi,
I’m just concerned if there is a specific reason why selective genes’ names are shown in the family merge taxonomy output, whereares in other higher level merges, only until the chosen levels were shown (below photos). But if the merged values should be correct, which means even though only 1 genus name under that family level was printed, still all the genes under the same family are merged, I could just pick out the family level names.
For example, below are class and order level merges:
Could you post the version of mothur you are using and the exact commands you ran? If you want to send your input files to mothur.westcott@gmail.com I can trouble shoot the issue for you.
Thank you for reporting this bug and providing your input files. I was able to fix the issue and the change will be part of 1.48.3 releasing next week. You can track the progress here.
