Unconfigured Ad

Collapse
X
 
  • Time
  • Show
Clear All
new posts
  • DNAjunk
    Member
    • Jun 2009
    • 62

    #1

    gsMapper: how to reduce number of gaps ?

    We would like to ask whether anyone else has seen the same problems when using 454's gsMapper with a set of reads and a reference sequence. How can one reduce the number of gaps that were introduced by the mapping alogrithm?

    Problem 1:
    In the ace file, one often sees gaps introduced by gsMapper at the begin of a read that could be perfectly aligend to the reference sequence. We think that this gap is a bug in the algorithm and should be avoided. Is there a gsMapper parameter that can be modified in order to prevent such gaps? Gaps are introduced only at the begin of a read, not at the end.

    HTML Code:
    read_1 GTCCGTTGCTTTAAGTGCTTGTTTATTCAT
    read_2 GTCCGTTGCTTTAAGTGCTTGTTTATTCAT
    read_3 GTCCGTTGCTTTAAGTGCTTGTTTATTCAT
    read_4 GTCCGTTGCTTTAAGTGCTTGTTTATTCAT
    read_5 GTCCGTTGCTTTAAGTGCTTGTTTATTCAT
    read_6 GTCCGTTGCTTTAAGTGCTTGTTTATTCAT
    read_7      TTGCTTTAAGTGCTTGTTTATTCAT
    read_8      TTGCTTTAAGTGCTTGTTTATTCAT
    read_9 GTCCGTTGCTTTAAGTGCTTGTTTATTCAT
    read_10  C.GTTGCTTTAAGTGCTTGTTTATTCAT
    read_11                     TT.ATTCAT
    read_12                         T.CAT
     
    contig GTCCGTTGCTTTAAGTGCTTGTTTATTCAT
    ref    GTCcGTTGCTTTAAGTGCTTGTTtATtCAT


    Problem 2:
    At another position in the ace file, gsMapper inserts gaps to avoid a mismatch. We want that the alignment has higher priority than inserting gaps. Is there a parameter in gsMapper that prevents such gaps? In the cases shown below, we prefer a mismatch than a gap.
    HTML Code:
    read_1  CC.AGAAACG.AGGAGACTCTTGGGA.TAGCGCCAAG.TCA
    read_2  CC.AGAAACG.AGGAGACTCTTGGGA.TAGCGCCAAG.TCA
    read_3  CC.AGAAACG.AGGAGACTCTTGGGA.TAGCGCCAAG.TCA
    read_4  CCN.GAAACGN.GGAGACTCTTGGGA.TAGCG 
    read_5  CC.AGAAACGN.GGAGACTCTTGGGA.TAGCGCCAAG.TCA
    read_6  CC.AGAA.CG.AGGAGACTCTTGGGA.TAGCGCCAAG.TCA
    read_7  CC.AGAAACG.AGGAGACTCTTGGGA.TAGCGCCAAG.TCA
    read_8  CC.AGAAACG.AGGAGA 
    read_9  CC.AGAAACG.AGGAGAC 
    read_10 CC.AGAAACG.AGGAGACTC 
    read_11 CC.AGAAACG.AGGAGACTCTTGGGA.TAGCGCCAAG.TCA
    read_12 CC.AGAAACG.AGGAGACTCTTGGGA.TAGCGCCAAG.TCA
    read_13 CC.AGAAACG.AGGAGACTCTTGGGA.TAGCGCCAAG.TCA
    read_14          G.AGGAGACTCTTGGGA.TAGCGCCAAG.TCA
    read_15            AGGAGACTCTTGGGA.TAGCGCCAAG.TCA
    read_16             G.AGACTCTTGGGA.TAGCGCCAAGC.CA
    read_17                  CTCTTGG.AATAGCGCCAAG.TCA
    read_18                            TAGCGCCAAG.TCA
    read_19                                   A.G.TCA
     
    contig  CC.AGAAACG.AGGAGACTCTTGGGA.TAGCGCCAAG.TCA
    ref     CC-aGAAaCG-aGgAGACTCTTGGgA-TAGCGCCAaG-tCA
    Many thanks for help!
  • maubp
    Peter (Biopython etc)
    • Jul 2009
    • 1544

    #2
    Problem 1 does look like a bug, probably related to how differences in homopolymer runs are normally handled in the middle of the alignment.

    I'm aware of other people grumbling about Problem 2, since this kind of output makes various downstream SNP finders unhappy. I'd have to ask around, but I suspect they took the practical way out and wrote a script to "fix" the alignment from Newbler.
    Last edited by maubp; 03-03-2010, 05:19 AM. Reason: typo

    Comment

    • maubp
      Peter (Biopython etc)
      • Jul 2009
      • 1544

      #3
      What version of gsAssember are you using? Reportedly "Problem 2" was fixed in Newbler 2.3 (something neither I nor the colleague I just spoke to have verified personally).

      Comment

      • DNAjunk
        Member
        • Jun 2009
        • 62

        #4
        Thanks for your help.

        Yes, we are using the latest version 2.3

        Comment

        • maubp
          Peter (Biopython etc)
          • Jul 2009
          • 1544

          #5
          Oh. I'm not in the office today, but I'll make a mental note to ask the other group more about this next time I see them.

          Comment

          Latest Articles

          Collapse

          • SEQadmin2
            Multiomic Liquid Biopsy: What DNA/RNA Panels Reveal That DNA Alone Can’t
            by SEQadmin2



            Tumors contain complex mixtures of cell populations that evolve in response to treatment and host-specific factors. Tissue biopsies provide only static snapshots of a tumor’s state. But what if we could follow that evolution over time using a minimally invasive technique?

            Liquid biopsies can detect tumor-derived material circulating in the bloodstream, providing a way to study treatment response, detect emerging resistance, and discover new biomarkers. Circulating
            ...
            10-14-2026, 09:55 AM
          • SEQadmin2
            New Genomics Technologies Take Aim at Long-Standing Limits
            by SEQadmin2


            Researchers using sequencing and genomics tools often have to make trade-offs. They can choose between speed or scale, short reads or long-range information, or targeted panels or a view of the whole transcriptome. New technologies that have been released this year are built to address those tough choices.

            We asked six companies the same four questions to learn about their latest products. The new technologies bring a lot to the table, including rethinking sequencing
            ...
            09-28-2026, 10:25 AM

          ad_right_rmr

          Collapse

          News

          Collapse

          Topics Statistics Last Post
          Started by SEQadmin2, 10-15-2026, 10:52 AM
          0 responses
          19 views
          0 reactions
          Last Post SEQadmin2  
          Started by SEQadmin2, 10-14-2026, 05:39 AM
          0 responses
          26 views
          0 reactions
          Last Post SEQadmin2  
          Started by SEQadmin2, 10-13-2026, 09:56 AM
          0 responses
          22 views
          0 reactions
          Last Post SEQadmin2  
          Started by SEQadmin2, 10-09-2026, 11:14 AM
          0 responses
          28 views
          0 reactions
          Last Post SEQadmin2  
          Working...