Unconfigured Ad

Collapse
X
 
  • Time
  • Show
Clear All
new posts
  • Antony03
    Member
    • Apr 2012
    • 53

    #1

    BWA - read identity

    Hi,

    I'm using BWA to map my short sequencing reads (Illumina MiSeq) on a reference sequence. Is this possible to set the minimum % of identity with BWA? For exemple, if I set 90%, then all reads under this threshold will be discarded. I know the parameter "-n" can do something like this but the manual is not really clear (for me). Is "-n" the minimal absolute mismatch in term of base pairs?

    Thank you!

    Antony
  • Brian Bushnell
    Super Moderator
    • Jan 2014
    • 2709

    #2
    Since nobody else has answered this, I will mention that I do not know of a way to generically do that in BWA. The manual says:
    -n NUM Maximum edit distance if the value is INT, or the fraction of missing alignments given 2% uniform base error rate if FLOAT. In the latter case, the maximum edit distance is automatically chosen for different read lengths. [0.04]
    ...which is clear in the case of int but unclear in the case of float. If all your reads are the same length you can calculate the edit distance that would correspond to that identity, though: e.g. 150bp * (1-0.90) = 15.

    Alternately, you can use BBMap, which has an "idfilter" flag - e.g. "idfilter=0.90", which will discard reads that don't map with at least 90% identity, regardless of their length. It also has an "idtag" flag, which will annotate all the alignments in the sam file with the percent identity, so you can manually filter them if you want.

    Comment

    • dariober
      Senior Member
      • May 2010
      • 311

      #3
      [This is a hack. I'm putting it here more for curiosity than anything else]

      On the aligned bam file you could capture the read length and the NM tag ("Edit distance to the reference, including ambiguous bases but excluding clipping") and print the alignment if L/NM < threshold. With samtools/awk:

      Code:
      samtools view -h aln.bam \
      | awk -v maxpct=0.05 -v OFS='\t' -v FS='\t' '{
          if($0 ~ /^@/){
              print $0
          } else {
              xnm=gensub(/.*\tNM:i:/, "", "g", $0); 
              nm=gensub(/\t.*/, "", "g", xnm); 
              pct=nm/length($10);
              if(pct < maxpct) {
                  print $0, "XP:f:"pct
              }
          }
      }' | less
      (The L/NM goes in tag XP)

      Comment

      Latest Articles

      Collapse

      • kinpict.com
        Reply to 16S/ITS Profiling vs. Shotgun Metagenomics: What Changes in Your Workflow?
        by kinpict.com
        The part about this topic feels surprisingly grounded, and the details on the page is the bit that made me slow down. I’d keep <a href="https://kinpict.com/">ai family photo generator</a> bookmarked for later.
        Today, 06:37 PM
      • SEQadmin2
        16S/ITS Profiling vs. Shotgun Metagenomics: What Changes in Your Workflow?
        by SEQadmin2





        NGS has revolutionized clinical microbiology by offering new ways to identify and surveil pathogens without prior knowledge of the target. One popular approach is 16S rRNA and ITS amplicon sequencing, which targets specific microbial markers. But studies are increasingly incorporating shotgun metagenomics, a technique that sequences all DNA from a sample without targeting specific microbial markers. This approach allows more detailed taxonomic classification,...
        10-30-2026, 06:46 AM

      ad_right_rmr

      Collapse

      News

      Collapse

      Topics Statistics Last Post
      Started by SEQadmin2, 11-12-2026, 07:59 AM
      0 responses
      27 views
      0 reactions
      Last Post SEQadmin2  
      Started by SEQadmin2, 11-05-2026, 09:25 AM
      0 responses
      35 views
      0 reactions
      Last Post SEQadmin2  
      Started by SEQadmin2, 11-03-2026, 11:54 AM
      2 responses
      56 views
      0 reactions
      Last Post kinpict.com  
      Started by SEQadmin2, 10-28-2026, 09:57 AM
      0 responses
      175 views
      0 reactions
      Last Post SEQadmin2  
      Working...