Hi everybody, I was trying to use the BLAST+ consensus taxonomy classifier to run six 16S datasets and I was unsure about a couple of things. First, the running time seems to be quite high with the configurations I am using. Here my inputs: My query sequences are quite big with ~60 MB - I am not s…

Hi @steff1088 , Thanks for posting! (and thanks for checking out that pre-print first to compare runtime results and parameters) [image] steff1088: How do these configuration parameters influence the running times individually? I have not actually benchmarked the effect of individual paramet…

Hi @Nicholas_Bokulich , thanks for coming back to me with these very helpful comments. I basically spent the last days following up on your suggestions. The vsearch classifier seems to be a good option and I am running it currently on 5 threads (also using SILVA's 7 level taxonomy instead of all). T…

[image] steff1088: paired-end pre-joined sequences cannot be used in the qiime 2 workflow prior to 2017.11 is that right? They should not be passed to dada2, which needs to operate on quality scores from forward and reverse reads separately. deblur and OTU picking (with q2-vsearch) are fine, …

After dereplicating with vsearch dereplicate-sequences and closed-reference clustering by vsearch cluster-features-closed-reference (SILVA's 97_otus) at 97% identity I looked at the output tables of 3 sample sets (containing 10 samples respectively) which I am each using as input for the vsearch cl…

Wow, I haven't seen so many features since the last time I used OTU picking :wink:. dada2/deblur will not necessarily reduce the number of features (some users see an increase because these methods essentially detect 100% OTUs, minus erroneous sequences), but with this many features my guess is that…

That's correct, Deblur does not use PHRED scores but it is assumed the data have been quality filtered (e.g., q2-quality-filter). Deblur processes samples independently when --p-min-reads is set to 1; combining sample sets or processing the sample sets separately will produce an identical result. Th…

[image] steff1088: What would be the best way forward in order to account for the pre-merged status of my sequence data using the new 2017.11 release? The QIIME 2 2017.11 release has expanded support for analyzing paired end reads! See the paired end reads community tutorial for more details.…

Thank you very much for your support @Nicholas_Bokulich @wasade . The samples are 16S Illumina reads and I am running subsets to test the workflow first. I used qiime feature-table filter-features to remove doubletons and singletons and my sequence numbers shrunk drastically. Now, before I would …

Hi @steff1088 , Your workflow looks correct to me; there is also a community tutorial on OTU picking here that may be useful in case you have not already seen it. [image] steff1088: I am running subsets to test the workflow first :heart: [image] steff1088: Could I theoretically use taxa…

QIIME 2 2017.12 is now out and includes the ability to use a feature table for filtering with filter-seqs!

QIIME 2 Forum

Performance and running time of classify-consensus-blast+

User Support

taxonomy, feature-classifier, paired-end

show post in topic

Home
Categories
Guidelines
Terms of Service
Privacy Policy

Powered by Discourse, best viewed with JavaScript enabled