kneepoint not functioning when no errors found

[DanYamamotoEvans]

When correcting sequences for each element, the program detects a 'kneepoint' in order to extract abundant sequences. In barcodeCorrector.py, the sequences above the kneepoint threshold is retrieved using the kneepoint rank as [:kneepoint-1].
This is an issue when there is only one sequence of that element (due to low read counts, high quality sequencing, or both), where extracted sequences becomes None ( [:0] ).

[DanYamamotoEvans]

please see pull request #5 (comment)