In the Oases paper (http://bioinformatics.oxfordjournals.org/content/28/8/1086.full), the following paragraph occurs:
To reduce the amount of erroneous bases, both paired-end datasets were processed by (i) removing Ns from both ends, (ii) clipping bases with a Sanger quality ≤10 and (iii) removing reads with more than six bases with Sanger quality ≤10 after steps (i) and (ii), leading to a total of 30 940 088 and 64 441 708 reads for human and mouse, respectively.
I'm a bit confused as to what (ii) means. Any insight?
Thanks in advance!