Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
ngsfilte
#1
Hi all members,

We had used ION Torrent platform for Honey metabarcoding
We do not used any tags for PCR amplification purpose and sequenced pooling each barcode amplicons (viz. rbcL, matK, trnL, ITS).
In that case I am unable to separate sequences base on my particular marker.
What should we do for further analysis and is not possible to annotate unique sequences using ngsfilter without using forward tag and reverse tag(In our case Forward and reverse tags are not used).


Thank you,
Apurva Punvar
Gujarat
Reply
#2
Hi Apurva,

I have no access to a computer for checking my answer, but it should work like explain further.

First you have right, ngsfilter requires tags. This is a true limitation and as we are writing the new version of the obitools we will take care to it. In the mean time I propose you a workaround to be able to deal with your data.
In the NGS filter replace the tag columun by -:- indicating no tags on both the extremity. This is not really managed by ngsfilter but if you add the -u option to the ngsfilter command you will be able to indicate a file for storing all the erroneous sequences. In your case all the sequences will be considered as erroneous and will end in this file.

imagine the command :

    ngsfilter -t mysample.ngsfilter -u error.fastq input.fastq > output.fastq

The output.fastq file will be empty and all your sequences will be in error.fastq
Now you will rescue your sequence using obigrep

  obigrep -a ‘error:Cannot assign sequence to a sample’ error.fastq > good.fastq

Then you can add the sample name to each of the sequence

  obiannotate -S ‘sample:MySampleID’ good.fastq > sample.MySampleID.fastq

You redo the same thing or each sample then

  obiuniq -m sample sample.*.fastq > samples.uniq.fasta


All the best

Eric
Reply
#3
Thank you verymuch Eric
will try in that way hope it will work for me
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)