Seqanswers Leaderboard Ad

Collapse

Announcement

Collapse
No announcement yet.
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Blast raw score calculation

    Hello,

    I know this sounds strange but I am having problems to validate the BLAST raw alignment scores (not the bit scores). Here an example alignment:

    -------------------------------------------------------

    Query: 100009 (Length=100)

    Hit: NZ_ABBZ01004690 gi|153879981|ref|NZ_ABBZ01004690.1| Beggiatoa sp. PS, whole genome shotgun sequence
    Length=280

    Score = 176 bits (194), Expected = 6e-42
    Identities = 99/100 (99%), Gaps = 0/100 (0%)
    Strand=Plus/Minus

    TCTGTTCTGTTCCATTGATCTATATCTCTGTTTTGGTACCAGTACCATGCTGTTTTGGTTACTGTAGCCTTGTAGTATAGTTTGAAGTCAGGTAGCGTGA
    ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| ||||
    TCTGTTCTGTTCCATTGATCTATATCTCTGTTTTGGTACCAGTACCATGCTGTTTTGGTTACTGTAGCCTTGTAGTATAGTTTGAAGTCAGGTAGTGTGA

    -------------------------------------------------------

    My scoring matrix is:
    match: 2
    mismatch: -3
    gap open: 5
    gap extension: 2

    Following the formula from the NCBI web site http://www.ncbi.nlm.nih.gov/Educatio...t_Scores2.html (and what other aligners produce as raw score) this makes

    99*2 - 3 = 195 and not 194 like BLAST says:

    <Hsp_num>1</Hsp_num>
    <Hsp_bit-score>176.213077892943</Hsp_bit-score>
    <Hsp_score>194</Hsp_score>
    <Hsp_evalue>6.28580662796915e-42</Hsp_evalue>
    <Hsp_query-from>1</Hsp_query-from>
    <Hsp_query-to>100</Hsp_query-to>
    <Hsp_hit-from>108</Hsp_hit-from>
    <Hsp_hit-to>9</Hsp_hit-to>
    <Hsp_query-frame>1</Hsp_query-frame>
    <Hsp_hit-frame>-1</Hsp_hit-frame>
    <Hsp_identity>99</Hsp_identity>
    <Hsp_positive>99</Hsp_positive>
    <Hsp_gaps>0</Hsp_gaps>
    <Hsp_align-len>100</Hsp_align-len>

    In fact I noticed that all BLAST raw scores are even! I am using BLAST version 2.2.22+. Do I miss an import point here?

  • #2
    Did you try to use BLAST+ instead? If the problem does occur with BLAST+, I would switch. You might also want to update to the latest BLAST version.

    Comment


    • #3
      Yes, I am using BLAST+. I am currently running the latest version to see whether it makes any difference. Anyhow, this shoudn't be if it is really a bug. I couldn't find anything about it in the change logs from 2.2.22+ to 2.2.24+ so I rather think that it will give the same results with the latest version. What do you mean by "you should switch"?

      Comment


      • #4
        If the raw score is a huge issue for you, you might want to consider alternative alignment programs (switch to an alternative program).
        You could also try to write the NCBI people who wrote the BLAST+ paper.

        Comment


        • #5
          Yes thanks, I counterchecked with the latest BLAST+ version and found the same issue. The raw score is important to me because when I compare different aligners using the same scoring scheme should result in more or less the same alignments with the same raw scores.

          Another point is that BLAST bit scores are calculated from the raw scores. This would mean that also the bit scores are slightly wrong!

          I will write them an email, unfortunately they don't have a developers mailing list at the NCBI...

          Comment


          • #6
            Just to let you know: Here is some evidence on the issue. I checked different versions and engines of NCBI BLAST. This seems to be an optimization in the code that is on purpose. The effect is that raw score can differ up to 3 points from real score.

            blast2.2.24+:

            * * * <Parameters_expect>10000</Parameters_expect>
            * * * <Parameters_sc-match>2</Parameters_sc-match>
            * * * <Parameters_sc-mismatch>-3</Parameters_sc-mismatch>
            * * * <Parameters_gap-open>5</Parameters_gap-open>
            * * * <Parameters_gap-extend>2</Parameters_gap-extend>
            * * * <Parameters_filter>F</Parameters_filter>

            * * * * * * * <Hsp_num>1</Hsp_num>
            * * * * * * * <Hsp_bit-score>176.213077892943</Hsp_bit-score>
            * * * * * * * <Hsp_score>194</Hsp_score>
            * * * * * * * <Hsp_evalue>6.28580662796915e-42</Hsp_evalue>
            * * * * * * * <Hsp_query-from>1</Hsp_query-from>
            * * * * * * * <Hsp_query-to>100</Hsp_query-to>
            * * * * * * * <Hsp_hit-from>108</Hsp_hit-from>
            * * * * * * * <Hsp_hit-to>9</Hsp_hit-to>
            * * * * * * * <Hsp_query-frame>1</Hsp_query-frame>
            * * * * * * * <Hsp_hit-frame>-1</Hsp_hit-frame>
            * * * * * * * <Hsp_identity>99</Hsp_identity>
            * * * * * * * <Hsp_positive>99</Hsp_positive>
            * * * * * * * <Hsp_gaps>0</Hsp_gaps>
            * * * * * * * <Hsp_align-len>100</Hsp_align-len>
            * * * * * * *
            <Hsp_qseq>TCTGTTCTGTTCCATTGATCTATATCTCTGTTTTGGTACCAGTACCATGCTGTTTTGGTTACTGTAGCCTTGTAGTATAGTTTGAAGTCAGGTAGCGTGA</Hsp_qseq>
            * * * * * * *
            <Hsp_hseq>TCTGTTCTGTTCCATTGATCTATATCTCTGTTTTGGTACCAGTACCATGCTGTTTTGGTTACTGTAGCCTTGTAGTATAGTTTGAAGTCAGGTAGTGTGA</Hsp_hseq>
            * * * * * * *
            <Hsp_midline>||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||| ||||
            </Hsp_midline>

            ================================================================================
            blast2.2.24:

            * * * <Parameters_expect>10000</Parameters_expect>
            * * * <Parameters_sc-match>2</Parameters_sc-match>
            * * * <Parameters_sc-mismatch>-3</Parameters_sc-mismatch>
            * * * <Parameters_gap-open>5</Parameters_gap-open>
            * * * <Parameters_gap-extend>2</Parameters_gap-extend>
            * * * <Parameters_filter>F</Parameters_filter>

            * * * * * * * <Hsp_num>1</Hsp_num>
            * * * * * * * <Hsp_bit-score>176.213</Hsp_bit-score>
            * * * * * * * <Hsp_score>194</Hsp_score>
            * * * * * * * <Hsp_evalue>6.28581e-42</Hsp_evalue>
            * * * * * * * <Hsp_query-from>100</Hsp_query-from>
            * * * * * * * <Hsp_query-to>1</Hsp_query-to>
            * * * * * * * <Hsp_hit-from>9</Hsp_hit-from>
            * * * * * * * <Hsp_hit-to>108</Hsp_hit-to>
            * * * * * * * <Hsp_query-frame>1</Hsp_query-frame>
            * * * * * * * <Hsp_hit-frame>-1</Hsp_hit-frame>
            * * * * * * * <Hsp_identity>99</Hsp_identity>
            * * * * * * * <Hsp_positive>99</Hsp_positive>
            * * * * * * * <Hsp_align-len>100</Hsp_align-len>
            * * * * * * *
            <Hsp_qseq>TCACGCTACCTGACTTCAAACTATACTACAAGGCTACAGTAACCAAAACAGCATGGTACTGGTACCAAAACAGAGATATAGATCAATGGAACAGAACAGA</Hsp_qseq>
            * * * * * * *
            <Hsp_hseq>TCACACTACCTGACTTCAAACTATACTACAAGGCTACAGTAACCAAAACAGCATGGTACTGGTACCAAAACAGAGATATAGATCAATGGAACAGAACAGA</Hsp_hseq>
            * * * * * * *
            <Hsp_midline>|||| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
            </Hsp_midline>

            ================================================================================

            blast2.2.24-legacy:

            * * * <Parameters_expect>10000</Parameters_expect>
            * * * <Parameters_sc-match>2</Parameters_sc-match>
            * * * <Parameters_sc-mismatch>-3</Parameters_sc-mismatch>
            * * * <Parameters_gap-open>5</Parameters_gap-open>
            * * * <Parameters_gap-extend>2</Parameters_gap-extend>
            * * * <Parameters_filter>F</Parameters_filter>

            * * * * * * * <Hsp_num>1</Hsp_num>
            * * * * * * * <Hsp_bit-score>177.115</Hsp_bit-score>
            * * * * * * * <Hsp_score>195</Hsp_score>
            * * * * * * * <Hsp_evalue>3.36455e-42</Hsp_evalue>
            * * * * * * * <Hsp_query-from>100</Hsp_query-from>
            * * * * * * * <Hsp_query-to>1</Hsp_query-to>
            * * * * * * * <Hsp_hit-from>9</Hsp_hit-from>
            * * * * * * * <Hsp_hit-to>108</Hsp_hit-to>
            * * * * * * * <Hsp_query-frame>1</Hsp_query-frame>
            * * * * * * * <Hsp_hit-frame>-1</Hsp_hit-frame>
            * * * * * * * <Hsp_identity>99</Hsp_identity>
            * * * * * * * <Hsp_positive>99</Hsp_positive>
            * * * * * * * <Hsp_align-len>100</Hsp_align-len>
            * * * * * * *
            <Hsp_qseq>TCACGCTACCTGACTTCAAACTATACTACAAGGCTACAGTAACCAAAACAGCATGGTACTGGTACCAAAACAGAGATATAGATCAATGGAACAGAACAGA</Hsp_qseq>
            * * * * * * *
            <Hsp_hseq>TCACACTACCTGACTTCAAACTATACTACAAGGCTACAGTAACCAAAACAGCATGGTACTGGTACCAAAACAGAGATATAGATCAATGGAACAGAACAGA</Hsp_hseq>
            * * * * * * *
            <Hsp_midline>|||| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
            </Hsp_midline>

            Comment

            Latest Articles

            Collapse

            • seqadmin
              The Impact of AI in Genomic Medicine
              by seqadmin



              Article Coming Soon......
              Today, 02:07 PM
            • seqadmin
              Multiomics Techniques Advancing Disease Research
              by seqadmin


              New and advanced multiomics tools and technologies have opened new avenues of research and markedly enhanced various disciplines such as disease research and precision medicine1. The practice of merging diverse data from various ‘omes increasingly provides a more holistic understanding of biological systems. As Maddison Masaeli, Co-Founder and CEO at Deepcell, aptly noted, “You can't explain biology in its complex form with one modality.”

              A major leap in the field has
              ...
              02-08-2024, 06:33 AM

            ad_right_rmr

            Collapse

            News

            Collapse

            Topics Statistics Last Post
            Started by seqadmin, 02-23-2024, 04:11 PM
            0 responses
            31 views
            0 likes
            Last Post seqadmin  
            Started by seqadmin, 02-21-2024, 08:52 AM
            0 responses
            46 views
            0 likes
            Last Post seqadmin  
            Started by seqadmin, 02-20-2024, 08:57 AM
            0 responses
            37 views
            0 likes
            Last Post seqadmin  
            Started by seqadmin, 02-14-2024, 09:19 AM
            0 responses
            63 views
            0 likes
            Last Post seqadmin  
            Working...
            X