Hi Haluk,
It is definitely not conventional to remove bases that have low quality,
because this disrupts the DNA sequence and may introduce frameshifts. It
is typically better to trim the ends of the sequence, or to remove it
altogether if its quality does not match your requirements.
Regards,
Florent
On 31/07/11 18:40, Haluk Dogan wrote:
Hi Florent,
I actually want to do removing respected bases if their quality score
is less than my threshold.
I had been able to do masking those bases by using FASTQ Masker by
quality score tool.
But I haven't gone even one step further for removing bases.
After your reply, I got suspicious. Am I trying to do something wrong
in theory?
Thanks in advance.
On Sun, Jul 31, 2011 at 6:07 AM, Florent Angly
<florent.angly(a)gmail.com <mailto:florent.angly@gmail.com>> wrote:
Hi Haluk,
By filtering, you mean removing reads? trimming their ends? or
masking some of their bases?
There are 3 tools under "Generic FASTQ manipulation" that may help
you:
Filter FASTQ reads by quality score and length
FASTQ Quality Trimmer by sliding window
FASTQ Masker by quality score
Regards,
Florent
On 31/07/11 05:58, Haluk Dogan wrote:
> Hi,
>
> I am trying to filter my fastq file with the condition of if
> quality score of reads is less then min score.
>
> So far, I have tried both *fastq_quality_filter* and *Filter
> FASTQ under NGS: QC and manipulation** *but I was not be able to
> do it.
>
> In the following you can see my fastq file.
>
> @F4HZV5G02CX6WP rank=0000096 x=1092.0 y=1767.0 length=45
> TTGAGCAGCGGCGTCACGGCGGCGGCCTCGGCGGCCGCATAGGCG
> +
> FFFFFFFFFFFIIIIIIIIIIIIIIIIIIIIIIIIIHFFDDBDA>
>
> And these are quality scores.
> [37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 37, 40, 40, 40, 40, 40,
> 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40,
> 40, 40, 40, 40, 39, 37, 37, 35, 35, 33, 35, 32, 29]
>
> I want to filter bases if their quality scores are less than 33.
>
> Any help would be greatly appreciated.
>
> --
> HD
>
>
>
> ___________________________________________________________
> The Galaxy User list should be used for the discussion of
> Galaxy analysis and other features on the public server
>
atusegalaxy.org <
http://usegalaxy.org>. Please keep all replies on the
list by
> using "reply all" in your mail client. For discussion of
> local Galaxy instances and the Galaxy source code, please
> use the Galaxy Development list:
>
>
http://lists.bx.psu.edu/listinfo/galaxy-dev
>
> To manage your subscriptions to this and other Galaxy lists,
> please use the interface at:
>
>
http://lists.bx.psu.edu/
--
HD