I'm try to use Qiime with RTAX assign_taxonomy method to deal with miseq PE data, which is not sequenced_through the V3 region.
but the result turn out to nothing assigned!
I follow the Qiime tutorial http://qiime.org/tutorials/rtax.html.
and here is my code :
assign_taxonomy.py -i leftseqs.fna_rep_set.fasta -m rtax --read_1_seqs_fp left/leftseqs.fna --read_2_seqs_fp right/rightseqs.fna -r 97_otus.fasta -t 97_otu_taxonomy.txt
also here is the repsentative sequence I use and the log file when run with RTAX.
the fa file is too big for upload.
thanks so much
I have made a blastn against nt database with some of your sequences (10) and all of them have a percent of identity below 97 or match with an "Uncultured bacterium" that could be no present in your database. Maybe you need to decrease this cutoff (default on qiime) to 94 or 91
And don't worry about the region sequenced because the greengenes database used by qiime includes the entire sequence of 16S rRNA (~1500bp).
Good luck!!!
@José Francisco Cobo Díaz
thanks
but my result show no sequence assigned. and the RTAX log file show that the can't find the paired_reads for my reperentative sequence to my OTUs.
when I use uclust method for taxonomy_assigned, I can use only the forward sequence to get the result. but when I want use the reverse sequence with the RTAX, the whole result is empty.
Do you know why this happened?
Sometimes the reverse sequences don't have enough quality to pass the quality filters by default from Qiime or another software. It is really a technical problem with the Miseq reagents, and there are some publications where they have had the same problem and finally they work only with the forward sequences.
Hi Junjiang,
It might worth trying UPARSE method that is superior than that implemented in QIIME.
I recommend you read the UPARSE puplication found at:
http://www.nature.com/nmeth/journal/v10/n10/full/nmeth.2604.html
We have build a webserver ( www.imngs.org ) that among others offers an implementation of UPARSE (Analysis tab) so it is super easy to have a try with your paired-end reads. If you register, this will give you an OTU table with few clicks in an hour or so. Then you can finish the downstream analysis using Rhea, our R based workflow for amplicon analysis available at:
https://lagkouvardos.github.io/Rhea/
if your reads are already demultiplexed into samples use the Perl script available at IMNGS to merge them back to a study and then run them through IMNGS.
If you have any problem with the procedure feel free to ask.
All the best with your project,
Ilias
www.imngs.org
- Badges
- Science topic
- Similar topics
- Bioinformatics
- Bioinformatics and Computational Biology