Bowtie - SOLiD : Weird decoding examples

[smarthey]

Hi,
I listed below some differents and weird mapping results obtain with Bowtie 0.12.7 on solid data.
1: SNP appearance

2_18_1468_F3    0    3    19908716    255    22M    *    0    0    ACCACAGGGTAGAACCACGGTC    IqqqqqqqqqqqqqqqqqqqqI    XA:i:2    MD:Z:20A1    NM:i:1    CM:i:2

[1]-----Read: T3101112001322010113012
[2]--Decoded: 3101112001322010113012
[3]Reference: 3101112001322010113012
It's EXACTLY the same ! But the final decoded nucleotide sequence makes appear a A which is neither in the read sequence nor in the reference sequence.
What happens here ?...

2: Color appearance

2_28_457_F3    16    1    32349243    255    22M    *    0    0    CATGTCCTGCTGAATGTCTAAC    Iqq!!qqqqqqqqqqqqqqqqI    XA:i:2    MD:Z:3C18    NM:i:1    CM:i:2

[1]------Read: 1310202132120311223010T (reverse)
[2]--Decoded: 131120213212031122301
[3]Reference: 132220213212031122301
How can I observe three different color ? The decoded nuc shouldn't be in the read or in the reference ?

3: ????

2_28_457_F3    16    3    30644442    255    22M    *    0    0    CATGGAAGTAGTCCGTGAGCCA    IqqqqqqqqqqqqqqqqqqqqI    XA:i:2    MD:Z:17G0A3    NM:i:2    CM:i:2

[1]------Read: 1310202132120311223010T (reverse)
[2]--Decoded: 131020213212031122301
[3]Reference: 131020213212031102101

Is someone obtain the same kind of results ? Is it normal ? Maybe I missunderstoud the decoding colorspace step.

[1] Original read color sequence
[2] Decoded color sequence (from decoded nucleotide sequence)
[3] Reference sequence color: