I'm new to the bioinformatics field but reading QIIME2 documents and forum I'm slowly succeding in understanding what to do. In order to master this knowleadge and help some other new self-taught bioinformatic, I would ask here some of my doubts hoping in your usual help. I thank you anyway.
There are my questions:
what to do if the total reads passing the primer trim are very few despite writing the correct pair of primers?
adding "-p--match-read-wildcard" to trim primer comand sometime helps to increase the total of trimmed reads... but does it mean that the quality of work is reduced due to incertainity of that base calling in some reads? Is it better to leave those reads discarded?
a percentage of reads mearged after DADA2 about 30-50% it's acceptable if the quality of reads it's very low?
can I use --p-perc-identity 0.97 with SILVA-99% database to classify the rep-seqs?
why sometime very few rep-seqs (about 30%) are found in database after classification step? Can a classifier trained on my reads improve this result?
Thank you so much for your time