Help me find a trimmer

[danbrami]

Hi folks,

I need a command line trimmer that will handle a seq/qual pair or ab1 file (for sanger reads obviously).
I have tried Lucy and unfortunately, I found:
- it did not trim the resulting sequence, only given the clear range positions in header
-was not aggressive enough in removing low quality, ambiguous and vector contaminants.

I am trying TrimSeq but I am not sure it will do the job (I only want to look fo a specific vector, not a slew of them nor contaminants)

Ideally, you would direct to something like an emboss tool that does what i want quickly.

What do you folks use?

Thanks!

daniel

[simonandrews]

The trimmers for Sanger data generally didn't create modified versions of the file, since the ab1/scf etc files were pretty big binary files and it didn't make much sense to duplicate them. The more usual approach was what you mentioned which was to simply write out the range of usable sequence into a separate configuration file which was then read by the assembler to show which parts of the sequence to use. This also had the advantage that since the trimmed sequence was only masked you could always bring it back later on if it proved to be useful in your alignment.

For this kind of stuff we still use the pregap program which is part of the staden suite. This has all of the options you need for trimming either poor quality sequence or vector and if you want it can write out the good quality sequence to a fasta file (but not to a trace file) if you're not intending to take the sequences forward for assembly.

[Himalaya]

Hi folks,

I need a command line trimmer that will handle a seq/qual pair or ab1 file (for sanger reads obviously).
I have tried Lucy and unfortunately, I found:
- it did not trim the resulting sequence, only given the clear range positions in header
-was not aggressive enough in removing low quality, ambiguous and vector contaminants.

I am trying TrimSeq but I am not sure it will do the job (I only want to look fo a specific vector, not a slew of them nor contaminants)

Ideally, you would direct to something like an emboss tool that does what i want quickly.

What do you folks use?

Thanks!

daniel

Try QTrim (hiv.sanbi.ac.za/software/qtrim)
It is python script and is commandline. It only does quality trimming and gives you the clean reads. No any vector trimming or adaptor trimming. Input files must be fastq file or fasta/qual files combined. The quality score should be PHRED score based. The command I tried was:
python qtrim.py -fastq myfastqfile -m 30 -l 50 -o myoutputfile

[carmeyeii]

Himalaya,

I don't quite understand ehat the INTERCRAP parameter in the Qtrim script is for. Do you?

Thanks,
Carmen

[Himalaya]

Himalaya,

I don't quite understand ehat the INTERCRAP parameter in the Qtrim script is for. Do you?

Thanks,
Carmen

Carmen
Sorry i don't get which parameter you saying about. It works so easy for me. If you input file has 454 short reads in fastq format, the command i posted earlier works easily for you as well. Let me know for any help if you struggling.

Himalaya

[a_mt]

how about NGS QC Toolkit.


Implemented in perl.

[Himalaya]

how about NGS QC Toolkit.


Implemented in perl.

Please check the NGS QC toolkit paper (http://www.plosone.org/article/info:...l.pone.0030619). The QC steps are clear in it. Depending upon your sequence reads and trimming requirement, you should be able to decide whether NGS QC toolkit is good enough for trimming or not. NGS QC is not my choice. You can also run the same sequence file into different quality trimming programs to know which one is performing better quality control in terms of number of sequences output, average read length of the output reads and also the number of bases with quality score below your threshold.