calibrated quality string "B" full

[Trudy]

Hello everybody!

I have a question about the calibrated quality string in qseq.txt file.

If I have full called sequence BUT a calibrated quality string full of “B” and the read passes the filtering (1), I should discard this read (although passed the filter)?

And if I have two reads in which the first read didn’t passed the filter while the second read passed BUT the number of “B” within is similar,
which read should I discard? Otherwise I could cut the read after “B” appeared?

In a few words, how I must to consider the “B” score for the next anlysis??

[maubp]

Are you aware that trailing "B" qualities can have a special meaning - The Read Segment Quality Control Indicator? See this thread:

http://seqanswers.com/forums/showpost.php?p=17491&postcount=3

I would therefore trim off any trailing bases with a "B" quality, e.g.

http://news.open-bio.org/news/2010/04/illumina-q2-trim-fastq/

[tomc]

If it helps, here is a shell script to truncate reads at the first B score


fastq_trim_B.awk

#! /usr/bin/nawk -f
# tomc  
# trim trailing 'B' scores (and sequence) from Ilumina FastQ reads

/^@.*/     {FQ["DL"]=$0; next}
/^[ACTGN]/ {FQ["SQ"]=$0; next}
/^\+.*/    {FQ["QL"]=$0; next}
/^[B-h]/   {
  n=match($0,/B/);
  if(n) {n--} else {n=length($0)};
  printf("%s\n%s\n%s\n%s\n",
	FQ["DL"],substr(FQ["SQ"],1,n),
	FQ["QL"],substr($0,1,n));
  next
}