Illumina sequencing Quality error

[huangjun]

Hi, everyone!
Recently, i received the RNA-seq data, when i use FASTX_TOOKIT preprocessing the data, it shows that my data invalid, quality < 0; Is
there anyone encountered this problem before? thank you!

[ulz_peter]

Sounds to me as if FastX is expecting different quality encoding than you've got in your samples. Which platform does your data come from? What did you want to do with the FastX Toolkit? (I must admit I am not too experienced with that toolkit)

[gringer]

append the '-Q 33' command to fastX tools to allow them to work with quality scores that have a base of 33. This is undocumented in the command-line help for the tools, but should work with all tools that demand particular quality values.

[huangjun]

my data come from Illumina Hiseq 2000 , FASTX_Toolkit can trim the low quality bases, cut off the low quality bases and mask the reads.

[huangjun]

Originally Posted by gringer

append the '-Q 33' command to fastX tools to allow them to work with quality scores that have a base of 33. This is undocumented in the command-line help for the tools, but should work with all tools that demand particular quality values.

Hi, gringer, in my reads , if there have more than 1 bases quality lower than 0 , or so ito large amount of reads, my be we can't know which parameter to append ?

[gringer]

Hi, gringer, in my reads , if there have more than 1 bases quality lower than 0 , or so ito large amount of reads, my be we can't know which parameter to append ?

My experience with fastx toolkit is that it doesn't support quality values lower than 0. If you have any in there, then you have to use -Q 33, and make the adjustment manually.

[huangjun]

Hi, gringer. I want to know "make the adjustment msnuslly" means? thank you !

[gringer]

Here's the table from Wikipedia:

  SSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSS.....................................................
  ..........................XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX......................
  ...............................IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII......................
  .................................JJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJJ......................
  LLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLL....................................................
  !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~
  |                         |    |        |                              |                     |
 33                        59   64       73                            104                   126

 S - Sanger        Phred+33,  raw reads typically (0, 40)
 X - Solexa        Solexa+64, raw reads typically (-5, 40)
 I - Illumina 1.3+ Phred+64,  raw reads typically (0, 40)
 J - Illumina 1.5+ Phred+64,  raw reads typically (3, 40)
    with 0=unused, 1=unused, 2=Read Segment Quality Control Indicator (bold) 
    (Note: See discussion above).
 L - Illumina 1.8+ Phred+33,  raw reads typically (0, 41)

The only negative reads here are the old Solexa reads. These start at base 59, so you could either use 59 as a base for the fastx-toolkit and add 5 to whatever you use for quality thresholds (e.g. mask at q25, rather than q20), or use 33 as a base for everything and add the difference (e.g. mask at q51, rather than q20). This mental arithmetic is what I mean by doing it manually.