Problem in interprating warnings and results after using qiime picrust2 full-pipeline

37 views

Skip to first unread message

Rashmi Ira

unread,

Jan 18, 2024, 2:13:14 AMJan 18

to picrust-users

Hi,

I am performing PICRUSt analysis for a very first time, trying to use q2-picrust2 with my environmental samples (sludge samples).
Here is command I have used:

qiime picrust2 full-pipeline \

--i-table merged-table_EF_RW.qza
--i-seq merged-rep-seqs_EF_RW.qza
--output-dir q2-picrust2/q2-picrust2_output
--p-placement-tool epa-ng
--p-threads 32
--p-hsp-method mp
--p-max-nsti 2
--verbose

In response of the above command I was getting this warning.

Warning - 2 input sequences aligned poorly to reference sequences (--min_align option specified a minimum proportion of 0.8 aligning to reference sequences). These input sequences will not be placed and will be excluded from downstream steps.

This is the set of poorly aligned input sequences to be excluded: 8fad6efc85152fc02447ff4a180b8ad9, 02b86a50afd72eee48f6669b5aa67fa9

What it means????Can I use these results?

**43 of 3158 ASVs were above the max NSTI cut-off of 2.0 and were removed from the downstream analyses.

Please suggest whether, these values are of positive interest or not??? If not, then how can I play with parameter **

--p-max-nsti 2

Any suggestion or comment will be appreciated.

Best Regards,
Rashmi

Robyn Wright

unread,

Jan 22, 2024, 1:36:51 PMJan 22

to picrust-users

Hi there,

That means that two of your ASVs (those listed as being excluded) were very different to those included in the PICRUSt2 reference files, and therefore predictions can't be made for them.

The second part of the results show that 43 of your ASVs were above the maximum NSTI value of 2 that you gave. You can read more about the NSTI here: https://github.com/picrust/picrust2/wiki/Key-Limitations

The NSTI that you use will depend on your study and the level of confidence you want to have in the PICRUSt2 predictions given.

This means that a small number of the ASVs that were in your initial samples have not been included in the PICRUSt2 functional predictions. This is not typically a problem and is just a limitation of using functional predictions rather than e.g. shotgun sequencing of your samples.