Does dada2 keep read1 when it is not possible to merge paired end data? I am losing 70% of my reads when working with dada2 and am curious if it is partially due to dada2 not retaining either read1 or read2 when there isn’t sufficient overlap of sufficient quality to merge them.
I didn’t see this answered anywhere else but forgive me if it was.
Hi @jpetteng,
Unfortunately not, DADA2 simply drops any reads that fail to meet its requirements, whether it is at the filtering step or merging step. Have a look through your dada2-stats output to see where you are losing your reads if you are not sure where this is happening.