Hello Pat,
I build my OTU table and classify my seqs. I would like to make a taxa summary plot (bar chart or pie chart). I want to show how the percentage of my families or orders in each sample.
Is it possible to make this kind of chart in mothur.
Ben
The easiest way I’ve found is to just open the summary file in Excel and graph from there. If you include your groups file when you do the classification the summary gives you the total reads/OTUs per classification, as well as how many were in each group.
The only thing to watch out for is that you’ll need to filter the table to only show the taxonomic depth you desire, which is easily done in Excel or OpenOffice using the taxlevel column. For example, if I classify an E. coli sequence, the summary will look something like (apologies for formatting):
taxlevel taxon total
0 Root 1
1 Bacteria 1
2 Proteobacteria 1
3 Gammaproteobacteria 1
4 Enterobacteriales 1
5 Enterobacteriaceae 1
6 Escherichia 1
If graphed straight away, it would count each line as a new data point and I would end up with 7 observed sequences.
Hi, do you mean filter it in excel.
What does the “summary file” end with? “.summary”. I can’t find the file.
Yes, filter it.
When you use the classify commands you should get two outputs, a .tax file that has each sequence name and the taxonomy, and a .summary file which is what I mean. If it doesn’t take you too long, you could re-run the classify command if you can’t find it.
Thank you. I have one more question about making taxa summary plot.
What dataset do you use? You know, I believe most of people use the subsampled dataset to do alpha or beta analysis. In terms of taxa summary plots, do you use the total dataset or the subsampled data?
Ben
It probably doesn’t matter, but we use the subsample