Announcement

Collapse
No announcement yet.

similarity of the 16 HAs

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • similarity of the 16 HAs

    Code:
    [FONT=Fixedsys]phylogenetic similarity between the 16 most recently published
    amino-acid HA-sequences, one for each of the 16 HA-subtypes:
     
     
       |01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16
    ---+-----------------------------------------------
    01 |.. 67 42 42 64 62 42 49 51 44 55 47 48 40 40 48 
    02 |67    41 39 73 61 41 48 50 42 53 46 49 39 40 46 
    03 |42 41 -- 66 40 41 48 40 38 50 40 41 39 65 47 37 
    04 |42 39 66 -- 40 42 47 42 40 48 41 41 41 77 48 40 
    05 |64 73 40 40 -- 60 40 48 48 41 55 46 47 40 39 47 
    06 |62 61 41 42 60 -- 40 51 51 41 55 49 48 42 39 48 
    07 |42 41 48 47 40 40 -- 37 40 64 39 41 36 47 78 35 
    08 |49 48 40 42 48 51 37 -- 61 41 48 65 48 41 38 47 
    09 |51 50 38 40 48 51 40 61 -- 40 50 62 51 38 40 49 
    10 |44 42 50 48 41 41 64 41 40 -- 42 41 39 49 65 38 
    11 |55 53 40 41 55 55 39 48 50 42 -- 49 59 41 40 58 
    12 |47 46 41 41 46 49 41 65 62 41 49 -- 50 40 43 48 
    13 |48 49 39 41 47 48 36 48 51 39 59 50 -- 42 38 80 
    14 |40 39 65 77 40 42 47 41 38 49 41 40 42 -- 49 39 
    15 |40 40 47 48 39 39 78 38 40 65 40 43 38 49 -- 36 
    16 |48 46 37 40 47 48 35 47 49 38 58 48 80 39 36 -- 
     
     
     
    and the same for the 9 NAs:
     
      | 1  2  3  4  5  6  7  8  9
    --+--------------------------
    1 |-- 43 44 67 52 46 41 54 44 
    2 |43 -- 50 41 41 45 44 44 44 
    3 |44 50 -- 41 41 46 44 41 44 
    4 |67 41 41 -- 54 41 39 55 37 
    5 |52 41 41 54 -- 41 40 67 35 
    6 |46 45 46 41 41 -- 57 43 64 
    7 |41 44 44 39 40 57 -- 40 57 
    8 |54 44 41 55 67 43 40 -- 38 
    9 |44 44 44 37 35 64 57 38 -- 
     
     
     
    [/FONT]

    H5 is most close to H2 and distant to H3
    Last edited by gsgs; October 8, 2006, 01:57 AM.
    I'm interested in expert panflu damage estimates
    my current links: http://bit.ly/hFI7H ILI-charts: http://bit.ly/CcRgT

  • #2
    Re: similarity of the 16 HAs

    I count (at least) 13 different HA,9 different NA,2 different M2,
    2 different NS1. (25% net)

    Number of complete Influenza A amino-acid sequences in a 1%,2%,..,10% net are:

    PB2:250,70,24,12,7,5,4,2,2,1
    PB1:135,27,11,4,2,1,1,1,1,1
    PA:228,71,21,11,7,3,2,1,1,1
    HA:923,515,336,232,164,130,88,73,59,43
    NP:268,81,32,18,11,6,5,4,3,1
    NA:913,507,314,211,142,112,85,68,44,38
    M2:532,322,185,125,73,50,34,25,15,9
    M1:196,58,32,12,8,4,2,1,1,1
    NS2:317,131,78,45,19,14,12,9,6,5
    NS1:390,212,133,74,48,36,23,18,14,12


    20% net gives 16 HA but 12 NA, so I doubt that it's reasonable to distinguish 16 HA and 9 NA.

    PB1 shows the least variety, must have been coincidence that PB1 was replaced
    in the 1957 pandemic (along with HA,NA).
    Last edited by gsgs; October 10, 2006, 12:57 PM.
    I'm interested in expert panflu damage estimates
    my current links: http://bit.ly/hFI7H ILI-charts: http://bit.ly/CcRgT

    Comment


    • #3
      Re: similarity of the 16 HAs

      The similarities between the gene segments has little to do with reassortment. Reassortment does not chnage the sequence of gene segments. By definition, a reassortant has acquired a complete gene unchnaged in sequence.

      The sequence of the NS gene in Israel was recently reported. It was a Qinghai sequnece, and the identocal sequence was in an H9N2 isolate. Thus H5N1 and H9N2 had a Qinghai NS gene, and the H9N2 was a reassortant, but the gene was acquired intact, without any sequence change

      http://www.recombinomics.com/News/07...assortant.html

      The same think happened in India. The HA and NA were clearly Qinghai, but the M gene was virtually identical to a Clade 2 isolate from China

      http://www.recombinomics.com/News/08...assortant.html


      Your analysis doesn't really address reassorment. For H and N most of the 144 possible combinations have been identifed. These serotypes however are defined serologically, and do not use gene or protein sequences.

      Comment


      • #4
        Re: similarity of the 16 HAs

        yes, it's not really reassortment, but related.
        The number of different HA's or NA's determine the
        combinations, reassortants. I thought this were the best subforum.

        here are my 13*9*2*2 = 468 serotypes:

        -----------------------HA------------------
        A/pintail duck/Alberta/210/2002(H1N1))
        A/Berkeley/1/68 (H2N2))
        A/duck/Hong Kong/7/1975(H3N2))
        A/turkey/Minnesota/833/1980(H4N2))
        A/mallard/Netherlands/3/99(H5N2))
        A/shorebird/DE/12/2004(H6N8))
        A/ruddy turnstone/NJ/65/1985(H7N3))
        A/turkey/Ontario/6118/1968(H8N4))
        A/duck/Hong Kong/644/79(H9N2))
        A/bird/Germany/1949(H10N7)
        A/mallard/Xuyi/8/2004(H11N?))
        A/mallard duck/ALB/342/1983(H12N1))
        A/herring gull/DE/475/1986(H13N2))

        -----------------------NA------------------
        A/Chicken/Italy/1067/99(H7N1)
        A/duck/Kingmen/E322/04(H6N2)
        A/turkey/England/63(H7N3)
        A/laughing gull/DE/94/2000(H12N4)
        A/mallard duck/ALB/884/1984(H2N5)
        A/quail/Californa/1/90(H4N6)
        A/ruddy turnstone/DE/2764/1987(H10N7)
        A/domestic duck/Minnesota/1086/1980(H4N8)
        A/mallard duck/ALB/294/1977(H11N9)

        -----------------------M2------------------
        A/chicken/Hubei/wf/2002(H5N1))
        A/shorebird/Delaware/113/2001(H9N2))

        ----------------------NS1------------------
        A/duck/Hokkaido/Vac-1/04(H5N1))
        A/chicken/Italy/1082/99(H7N1))
        Last edited by gsgs; October 10, 2006, 01:08 PM.
        I'm interested in expert panflu damage estimates
        my current links: http://bit.ly/hFI7H ILI-charts: http://bit.ly/CcRgT

        Comment


        • #5
          Re: similarity of the 16 HAs

          Although the most similar hemagluttinin to H5 is H2, the second-most similar is H1. Worth noting, considering that Spanish Flu was an H1 virus.

          Comment


          • #6
            Re: similarity of the 16 HAs

            that was for nucleotide-differences, for protein-differences
            I get for 2%,3%,...,35% filter:
            591,364,249,170,131,100,73,62,50,
            40,30,28,24,22,22,21,21,19,16,
            16,14,14,13,13,12,12,12,12,12,
            12,12,12,12,11,

            HA-sequences.
            So distinguishing 12 HAs instead of 16 looks most nutural :
            H1,H3..13


            edit: oops, both are for protein sequences (amino-acids) not nucleotides.
            Database from October 2006.
            Last edited by gsgs; May 28, 2007, 02:52 PM. Reason: correction
            I'm interested in expert panflu damage estimates
            my current links: http://bit.ly/hFI7H ILI-charts: http://bit.ly/CcRgT

            Comment


            • #7
              Re: similarity of the 16 HAs

              ho gsgs, can you explain to me how read your first post ?
              I'm interesting
              thank you

              Comment


              • #8
                Re: similarity of the 16 HAs

                as far as I remember, I took one typical sequence from each
                type. A "central" one, one which is most similar to the rest
                in the group.
                Then I calculated the average number of nucleotide-differences
                of these 16 sequences, one for each HA-type.
                Numbers are the ratios of matching amino-acids divided by length,
                so 00 would be zero matches, 100 would be identical.

                To compare: I haven't checked this, but I assume that any two Qinghai-H5N1
                sequences would be 99 or 100
                I'm interested in expert panflu damage estimates
                my current links: http://bit.ly/hFI7H ILI-charts: http://bit.ly/CcRgT

                Comment


                • #9
                  Re: similarity of the 16 HAs

                  OK, merci beaucoup.

                  Comment


                  • #10
                    Re: similarity of the 16 HAs

                    I am confused. Relative to the above information and noting this quote from 'Breaking News,' I thought that a recombination of a LPAI such as H7N2 and the HPAI H5N1 would possibly be the final step to a pandemic strain.

                    However, it was stated that there was no evidence that a combination of H7/H5 would be bad. Then, what is bad? What are we looking at being the reassortment that could be the pandemic strain if this is not it? Is there a reason to be concerned about this current H7N2 flu?

                    Thank you.

                    Code:
                     Re: Confirmation of Avian Influenza H7N2 Infection 
                    
                    --------------------------------------------------------------------------------
                    
                    Quote:
                    Originally Posted by Vibrant62  
                    H7 viruses are also a concern, becuase, like H5 viruses they can convert into highly pathogenic forms quite readily, but not as readily as H5 viruses. H7 viruses will only become more pathogenic when additional amino acids are added to the cleavage site, whereas LPAI H5 viruses can become HPAI viruses by either an addition of an amino acid to the cleavage site or by substitution of an amino acid within the existing cleavage site.
                    
                    Just becase its not particularly 'bad' at present, does not mean it will remain that way.
                    
                    An H7N2 reassortment with H5N1 could be very bad indeed. 
                    
                    That is incorrect. Both H5 and H7 can convert by adding or changing amino acids at the cleavage site.
                    
                    There is no evidence that an H5/H7 reassortant would be bad.

                    Comment


                    • #11
                      Re: similarity of the 16 HAs

                      Originally posted by gjs47 View Post
                      I am confused. Relative to the above information and noting this quote from 'Breaking News,' I thought that a recombination of a LPAI such as H7N2 and the HPAI H5N1 would possibly be the final step to a pandemic strain.

                      However, it was stated that there was no evidence that a combination of H7/H5 would be bad. Then, what is bad? What are we looking at being the reassortment that could be the pandemic strain if this is not it? Is there a reason to be concerned about this current H7N2 flu?

                      Thank you.

                      Code:
                       Re: Confirmation of Avian Influenza H7N2 Infection 
                       
                      --------------------------------------------------------------------------------
                       
                      Quote:
                      Originally Posted by Vibrant62  
                      H7 viruses are also a concern, becuase, like H5 viruses they can convert into highly pathogenic forms quite readily, but not as readily as H5 viruses. H7 viruses will only become more pathogenic when additional amino acids are added to the cleavage site, whereas LPAI H5 viruses can become HPAI viruses by either an addition of an amino acid to the cleavage site or by substitution of an amino acid within the existing cleavage site.
                       
                      Just becase its not particularly 'bad' at present, does not mean it will remain that way.
                       
                      An H7N2 reassortment with H5N1 could be very bad indeed. 
                       
                      That is incorrect. Both H5 and H7 can convert by adding or changing amino acids at the cleavage site.
                       
                      There is no evidence that an H5/H7 reassortant would be bad.
                      Reassortment and recombination are quite different.

                      Comment


                      • #12
                        Re: similarity of the 16 HAs

                        Reassortment and recombination are quite different.
                        I have misunderstood the two terms. I have been understanding the two terms as being interchangeable - versus mutation as being an incorrect viewpoint on all fronts.

                        So, recombining is between two strains, whereas reassortment is a change within one strain. Is that correct?

                        Thank you, again.

                        Comment


                        • #13
                          Re: similarity of the 16 HAs

                          flu-A has 8 segments, when you have 2 different viruses in one cell,
                          they may permute the segments. That's reassortment. !957 and 1968
                          pandemics happened this way and the Qinghai-strain was created this way.

                          Sometimes it also happens that in one segment portions of 2 other
                          segments are used simultaneously. This is rarer (niman disagrees...)
                          and called recombination. It also requires similar strains, so a recombination
                          in HA from H5N1 and H7N2 looks unlikely.

                          H5N1 had already reassorted and (rarer) recombined with other
                          flu , i.e. H6N1 and H9N2 earlier and fetched some of their segments.

                          That way H5N1 was formed in the form as it exists now.
                          But only in China, AFAIK and no or few examples since 2004, I think.
                          I'm interested in expert panflu damage estimates
                          my current links: http://bit.ly/hFI7H ILI-charts: http://bit.ly/CcRgT

                          Comment


                          • #14
                            Re: similarity of the 16 HAs

                            taking the complete sequences from 07.March I get now
                            for PB2,PB1,PA and NP ,
                            filters 0%,0.1%,0.2%,0.3%,0.4%,0.5%,,
                            1%,2%,3%,4%,5%,6%,7%,8%,9%,10%

                            Code:
                            number of nucleotide-sequences in a (filter)% - grid:
                            PB2:3234,1809,1433,1176, 957, 862,,594,331,194,126,86,56,41,32,22,17
                            PB1:3258,1776,1412,1146, 953, 855,,572,309,168,113,72,39,24,18,14, 9
                            PA :3301,1751,1352,1141, 994, 888,,548,292,183,112,77,45,33,26,20,13
                            NP :3626,2076,1699,1310,1181,1001,,676,379,222,131,71,40,28,24,17,11


                            Code:
                            number of protein-sequences in a (filter)%-grid:
                            ----  0 ,0.1 ,0.2 ,0.3,0.4,0.5,, 1 , 2, 3, 4, 5,6,7,8,9,10 
                            ----------------------------------------------------------
                            PB2:3234,1601,1198,766,606,603,,232,57,24,10, 7,6,4,2,2,1
                            PB1:3258,1508,1102,664,520,511,,164,40,13, 4, 2,1,1,1,1,1
                            PA :3301,1606,1211,766,766,596,,227,78,22,14, 8,4,3,1,1,1
                            NP :3626,1290,1290,956,956,526,,283,89,35,20,12,6,5,3,2,2
                            so, diversity is smallest in PB1 - no flu-A sequence differs in more than
                            7% of its amino acids in PB1 from A/Azerbaijan/006-207/2006,(H5N1).

                            This makes PB1 the most likely candidate for reassortments and
                            this is confirmed by counting possible reassortment-pairs :
                            PB2+PB1:385086,245 candidates
                            PB2+PA :206639,144 candidates
                            PB2+NP : 82176,111 candidates
                            PB1+PA :374763,223 candidates
                            PB1+NP :462836,231 candidates
                            PA +NP : 99484, 98 candidates

                            (out of about 4.5million pairs)

                            so, when e.g. NP is acquired by reassortment then often PB2 and PA are acquired too,
                            while PB1 is more flexible to reassort in any combination.

                            PB1 also contains the 90-aa-long PB1-f2 protein, which is known to mutate much
                            on the protein level
                            Last edited by gsgs; August 29, 2007, 12:39 PM. Reason: list completed
                            I'm interested in expert panflu damage estimates
                            my current links: http://bit.ly/hFI7H ILI-charts: http://bit.ly/CcRgT

                            Comment

                            Working...
                            X