When quality of reads is too bad at 3'-end to allow for pair-end assembly, optimize truncation and just join R1 and R2 sequences with a padding sequence of NN's.
This will force the full analyses.
Clustering optimization will be needed (see #175)
Identification is done with vsearch iddef=0 (only matching columns) after removal of the padding sequence.
When quality of reads is too bad at 3'-end to allow for pair-end assembly, optimize truncation and just join R1 and R2 sequences with a padding sequence of NN's. This will force the full analyses. Clustering optimization will be needed (see #175) Identification is done with vsearch iddef=0 (only matching columns) after removal of the padding sequence.