biom format for PICRUST

Commands in mothur
1

Hi,

I have run my sequences with the latest mothur SOP. Some of my next analysis will be to generate metagenome data from my 16S sequences. My questions:

  1. I have aligned my data with silva. Will the following command be sufficient to get an output file that can be read/analyzed by PICRUST:

mothur > make.biom(shared=my_data.shared, label=0.03, reftaxonomy=gg_13_5_99.gg.tax, constaxonomy=my_data.0.03.cons.taxonomy, picrust=97.gg.otu_map)

  1. Is there any other program that takes the output of mothur (i.e. silva) to create metagenome data? Ideally, I’d like all my analysis with just one reference database (i.e. silva).

many thanks for your help!

2

Hi,

I’m pretty certain that to do picrust you have to use teh greengenes reference taxonomy.

Pat

3

Ok. So the commands (from mothur SOP) that need to be changed for green genes are:

  1. align.seqs(fast=current,reference=gg_13_8_99.fasta)
  2. screen.seqs(fasta=current, count=current, start=???, end=???, maxhomop=8)
  3. classify.seqs(fasta=current, count=current, reference=gg_13_8_99.fasta, taxonomy=gg_13_8_99.gg.tax, cutoff=80)
  4. make.biom(shared=abrecovery.an.shared, label=0.03, reftaxonomy=gg_13_8_99.gg.tax, constaxonomy=abrecovery.an.0.03.cons.taxonomy, picrust=97_otu_map)

Is this correct? What should the start and end positions be for screen.seqs()?

thanks for all your help!

4

seems to work with v1.39.5!

5

Not quite - you should never use greengenes to do an alignment. I would use steps 1 and 2 as listed in the MiSeq SOP, but do as you describe for 3 and 4.

Pat

6

Ah, yes, I read some other posts and only did steps 3 & 4.

Thanks for the help!