QIIME2 - database, sample integrity, cutoff scores

[colinbrislawn (Colin J Brislawn)]

Hello Aakarsha,

I'm glad you are making progress with Qiime2. Thanks for Being Patient until we had time to answer your questions.

@aakarsharao

• In the FeatureTable, if a sample shows a decreased frequency of features (or OTUs), does that mean the sample integrity has been compromised or sample may have been degraded during the initial sequencing steps?

If the sequencing frequency is near zero (10s or 100s of reads), then the sample integrity probably has been compromised during extraction or PCR, just like you suggested.

Keep in mind that on the Illumina platform, some samples sequence more deeply than others. We can try to get more even sequencing coverage by adding the same mass of PCR product from each sample, but I've never seen an Illumina run with 100% even reads per sample.

@aakarsharao

1. As I performed the steps of the "moving pictures" tutorial, I learned that there was only the Greengenes database that was used for the taxonomic profiling. Is there a way to choose a database of our liking while carrying out the QIIME pipeline?

Sure! In fact, the Moving Pictures tutorial includes a link to the Data Resources page that lists other database pre-made to work with Qiime2 including SILVA and Unite. You can also check out the RESCRIPt plugin for building and testing your own databases.

I hope this helps. Let us know if you have other questions,
Colin