Question: What Is The Best Way To Get Rid Of Low Quality Bases In A Sequence
0
gravatar for kmkdesilva
5.8 years ago by
kmkdesilva80
United States
kmkdesilva80 wrote:

Hi all,

I would like to know what is the best way to get rid of low quality bases in a sequence using fastx tool kit. Is quality filtering the sequence using a given phred score or trimming the sequence.

By quality filtering we might loose bases in the middle of the sequence. From trimming we can chop of the ends. So I would like to know what is the best way.

Regards, kmkdesilva

trimming fastx • 2.4k views
ADD COMMENTlink modified 5.8 years ago by rob234king580 • written 5.8 years ago by kmkdesilva80
0
gravatar for IV
5.8 years ago by
IV1.2k
USA
IV1.2k wrote:

Quality filtering by use of fastq_quality_filter does not clip bases within the sequence but rather rejects reads that do not meet the user defined criteria.

For instance you would want to keep reads that have more than q quality for at least p percentage of the read length.

fastq_quality_filter -q min_quality -p min_percent -i file -o outfile

You could select to trim when low quality base calls are at the last few nucleotides of your reads and to quality filter when there are more generalized quality issues.

In general, fastx toolkit tools are considered quite strict and you end up losing quite a lot of your sequences.

Other tools for trimming, adapter clipping and/or quality filtering are more gentle (such as scythe, cutadapt, etc).

Quality trimming and filtering should be performed with care, in order to actually increase the quality and quanity of what you get in the end.

Adapter removal to my opinion should be performed more vigorously.

Cheers,

IV

ADD COMMENTlink written 5.8 years ago by IV1.2k
0
gravatar for Haluk
5.8 years ago by
Haluk170
Lincoln, Nebraska
Haluk170 wrote:

I used the PoPoolation before, and it works well.

ADD COMMENTlink written 5.8 years ago by Haluk170
0
gravatar for rob234king
5.8 years ago by
rob234king580
UK/Harpenden/Rothamsted Research
rob234king580 wrote:

Primers and adaptors need to be watched out for, especially 454 data but the best trimming tends to be done by the software in question, atleast for mapping. For instance I dont trim using third party software with novoalign because it has a clipping function built in which I can use and when tested with my trimming it performs significantly better. Or sometimes even with all the junk will map better rather than trimming and just filter after. For assembly I found Quake good at automatically cleaning up reads and did show benefit. If I have to trim then trimmomatic is my favourite.

ADD COMMENTlink written 5.8 years ago by rob234king580
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 1842 users visited in the last hour