Optimal Alignment with Quality Rescoring
Rescoring Algorithm
- All recruited alignments were then optimally realigned using an in-house Needleman-Wunsch implementation (match = +10, mismatch = -8, gap opening = -20, gap extension = -1, no penalty for terminal gaps) (Needleman and Wunsch, 1970).
- Global realignment improves the treatment of insertion-deletion and substitutions.
- The percent identity for each global alignment was then recalculated, base base by including only those aligned bases where fosmid-end nucleotides were high quality.(phred Q score . 30, which equals a sequencing error rate of 10-3 per base, any bases with a Q score <30 is ignored)
Rescoring Parameter
- Global Alignment Score=basepairs * [2*identity-20 *[1-identity]]
- Alignments for each fosmid end were filtered to remove relatively small, lower scoring, low identity alignments which unlikely represent allelic representations