how quality is converted?

mothur bugs
1

I have a simple test.fastq file that contains just one sequence :

@I0V2C:4:4
GTGACCGGGATCGTTGTCCGCGAGCACGCTGCCGGACGGCGGTGCCTGTGCGACCGAGTACGACTGCCGAAGTCGAACTGGGCGGCGTCGACGAGGAACCGAGGCGGTACTGACGTAGACGGG
+
%-.66079-398±%-,1/8.::82/0,2-0%-(.440492437./:924228+577779;>>663063,3++/)1--%*1&,+7555;899/4,4.536.,&,06+,+////+±/7+

When I run fastq.info(fastq=test.fastq), the test.qual file is contains the following scores:

I0V2C:4:4
-27 -19 -18 -10 -10 -16 -9 -7 -19 -13 -7 -8 -21 -19 -27 -19 -20 -15 -17 -8 -18 -6 -6 -8 -14 -17 -16 -20 -20 -14 -19 -16 -27 -19 -24 -18 -12 -12 -16 -12 -7 -14 -12 -13 -9 -18 -17 -6 -7 -14 -12 -14 -14 -8 -21 -11 -9 -9 -9 -9 -7 -5 -2 -2 -10 -10 -13 -16 -10 -13 -20 -13 -21 -21 -22 -17 -23 -15 -19 -22 -19 -27 -22 -15 -26 -20 -21 -9 -11 -11 -11 -5 -8 -7 -7 -17 -12 -20 -12 -18 -11 -13 -10 -18 -20 -20 -26 -20 -16 -10 -21 -20 -21 -17 -17 -17 -17 -21 -21 -19 -17 -9 -21

These quality sccores are not correct…

2

Sorry about this - Illumina changed its method of calculating quality scores a while back and this is hte problem. We’ll get this fixed for you in the next release.

Pat

3

There are three variants of FASTQ format, solexa, illumina and sanger. It seems sanger FASTQ is becoming the standard FASTQ format. If fastq.info contains an option to specify these three variants with sanger variant as the default, it would be perfect.

4

Hello, I am having the same problem incorrect conversion of quality scores from illumina fastq files. Are you planning this for v 1.28? Do you have an estimate of when it might be available? Please know I’m not trying to be pushy at all, just trying to decide if I should figure out a workaround or simply wait for the version upgrade. Thanks!