#12 Maq Smith Waterman aligned deletions not shown

[Anthony Fejes]

When using PETs for alignments, the second sequence is often aligned using a Smith-Waterman algorithm, when the first is uniquely matched. In this case, gaps and deletions should be visible.

For the insertions (positive value for the 8th field), an extra base (or more) is observed.
SOLEXA8_4:3:83:975:590/2 chr10 3154466 - -219 130 32 1 69 2 50 0 0 42
agtgcggtggttctagtctcgtgatgggtcagacggcgccac

query: agtgcggtggttctagtctcgtgatgggtcagacggcgccac

genomic: AGTGCGGTGGTTCTAGTCTCGTGATGGGTGAGNTGGCGCCGTC

(Note position of N in genomic sequence corresponds with the insertion predicted)

For the deletions (negative value for the 8th field), the sequence provided matches the reference sequence, leading to the question of whether there is actually a deletion observed. (I can provide thousands of examples). Thus, either the sequence provided in the map file is no longer the same as in the query, or the alignment call is incorrect in the description of a deletion.

SOLEXA8_4:3:10:97:41/2 chr10 3155303 - -148 130 30 -2 65 3 65 0 0 42 cctccgggcagagccgaaggaatgagataccagagcacgtgt

Aligned against reference at that location:

query: cctccgggcagagccgaaggaatgagataccagagcacgtgt
genomic: cctccgggcagagccgaaggaatgagataccagagcacgtgt