A New Decade of ME Research: The 11th Invest in ME International ME Conference 2016
Mark Berry presents the first in a series of articles on the 11th Invest in ME International ME Conference in London ...
Discuss the article on the Forums.

Lo/Alter virus sequences

Discussion in 'XMRV Research and Replication Studies' started by jimm, Sep 2, 2010.

  1. jimm

    jimm

    Messages:
    22
    Likes:
    2
  2. George

    George waitin' fer rabbits

    Messages:
    851
    Likes:
    80
    South Texas
    Nice jimm.

    So the isolated 4 different gene sequences from CFS and 4 diffrent gene sequences from healthy diners (BD set) but state in the paper that there are only 6 unique virii. Can you tell which two match up?? I'm a bit foggy today.
     
  3. Otis

    Otis Señor Mumbler

    Messages:
    1,116
    Likes:
    116
    USA
    George - you had going with 'diners' instead of 'doners'. I'm not making fun, I substitute vowels (that aren't right next to each other on the keyboard!) to make new and interesting words all the time.
     
  4. George

    George waitin' fer rabbits

    Messages:
    851
    Likes:
    80
    South Texas
    egads, don't make me laugh I'm too tired today. Dang it! O.k. so they had 4 unique sequences from the CFSer's and 4 from um. . .Them healty people (grins) but only 6 unique virii so I'm wondering which ones match. Cause I'm thinking if the CFSer's match up that would me a specific type of virus for us and then I would think that those might be like really like the XMRV one.

    Then I'm thinkin' do the ones that show up in the healthy people look like the ones in the CFSer's??? Cause what if they kinda look like FMlV or MoMLV ya know where I'm going with this???

    I'm going to wander off an eat some doners now. . . . .
     
  5. jimm

    jimm

    Messages:
    22
    Likes:
    2
    I have been trying to research a bit on google about what implications this has but I have to admit defeat as its beyond my science knowledge....one thing I did find was a comment on a skeptic blog that claimed that when the data was "blasted" whatever that means! and that its possibly an endogenous MLV..and they said some evidence was that there was only two nucleotide differences......

    now the reason I mention this is I try to keep up with data both from skeptics and pro people, becuase it gives me a good overview....so I dont know if the claims of this skeptic are true or not and infact I have NO idea what they mean...but could someone in the know explain whether this claim of "ohhh they are endogenous MLV" holds any water?? anyone know or is this major techy stuff?
     
  6. Otis

    Otis Señor Mumbler

    Messages:
    1,116
    Likes:
    116
    USA

    So here is the list. It looks like #7 is an env gene sequence from a CFS patient (assuming that one because it's not got a BD #) and #8 is an env of a donor. Turns out this last env sequence is from the first pol donor (BD-26), so this list has only data from 3 unique blood donors, the 3 cfs types and envs from a donor and what I guess is a CFS patient.

    I played around with comparing these to one another with Blast. I'm still figuring out how it works. I've compared the first 6 and the shorter sequences appear to be contained within the longer ones aligning at BP 201 (of CFS type 1).

    A few examples from the first 6.
    CFS1 vs. CFS2 is only different in 2 BPs, 0 gaps.
    CFS1 vs. CFS3 has 9 bp differences and 2 gaps.
    CFS1 vs. BD-22 is different in 20 BPs and 1 gap.

    The two ENV sequences compare at 199 of 206 BPs the same with no gaps.

    George - I need a hint on the Friend and Moloney angle. My brain has melted down.
     
  7. alex3619

    alex3619 Senior Member

    Messages:
    12,528
    Likes:
    35,244
    Logan, Queensland, Australia
    Hi Jimm

    I think Otis might get what BLAST is (I think it is supposed to be all caps, its an acronym if I recall correctly). I haven't used it for a decade now, but it is a standard technique for searching through DNA/RNA databases looking for matches with a test sequence. So having sequenced a candidate virus, they run a BLAST search to find which viruses it is related to, and can locate exactly where it is different.

    Bye
    Alex
     
  8. Otis

    Otis Señor Mumbler

    Messages:
    1,116
    Likes:
    116
    USA
    I started playing around with it before I learned what it meant. :Retro smile:

    Here's the description from the NIH site:

    The Basic Local Alignment Search Tool (BLAST) finds regions of local similarity between sequences. The program compares nucleotide or protein sequences to sequence databases and calculates the statistical significance of matches. BLAST can be used to infer functional and evolutionary relationships between sequences as well as help identify members of gene families.
     
  9. Mark

    Mark Former CEO

    Messages:
    5,224
    Likes:
    6,197
    Sofa, UK
    OK I'm out of my depth I know, but I recollect that CFS1-3 are the 3 found in CFSers and the 4th they found was in those healthy diners (that's a classic btw!). Don't recall if there was any overlap, ie whether any of those were found in both. This may all be obvious but thought I'd post in case it helped...


    I'd like to know...I think...are you going towards bits of these sequences look like maybe they could be just those known 'harmless' MLVs you mention, or are you going towards hmm maybe those MLVs aren't so friendly as was thought? None of them can actually be FMIV or MoMLV surely, since they've said that all four are new PMRVs? Is the possibility of some of the new HGRVs being formed using bits of the harmless MLVs looking at all likely do you think?
     
  10. busybee

    busybee Senior Member

    Messages:
    119
    Likes:
    7
    I wish I could vote this thread HOT

    I love to see your brains unravelling all the pieces

    Thank you :D
     
  11. Otis

    Otis Señor Mumbler

    Messages:
    1,116
    Likes:
    116
    USA
    Two of these sequences have been updated!

    In the BD26 gag sequence there was a single bp insertion.

    In the BD28 gag sequence, however, there are significant differences.

    I see now that originally the BD26 gag and BD28 gag sequences were exactly the same.

    So they made a minor update to the former, and fixed a "copy/paste" error in the latter.

    It appears that the published phylogenetic tree was correct but until I learn how to find other sequences and make comparisons in BLAST I can't be sure.

    OK, me head hurts. :headache:

    George, I still need that hint on the Friend and Maloney thought.
     
  12. George

    George waitin' fer rabbits

    Messages:
    851
    Likes:
    80
    South Texas
    Yes my brain has completely unravelled and is now in pieces. (grins)

    O.k. now what the heck to do with the information that two of the sequences match 100 % to lab mouse 129 XMRV viral strain and one has a 96% sequence match to lab mouse C57 black/ C57 black 6 (same mouse line evidentaly) ?????????????????????????????????????????????????

    Otis my brain is in so many pieces at thsi ponit thta I fogret what the hekc I was thniking in the fisrt place. (bleh)
     
  13. Otis

    Otis Señor Mumbler

    Messages:
    1,116
    Likes:
    116
    USA
    George, it's cool. I know the feeling. Curl up with a nice bone, it always clears my mind. I'm guessing it was either a recombination theory (perhaps Sandra Ruscetti?) or a vaccine theory.

    I can run these 8 sequences against a mouse genetic DB in BLAST but I get so many hits I can't make any sense of the results. I think it's the short length of the of Lo/Alter sequences - the results line up matched, or very close to, many, many sequences. I'm a beginner at this and I think I need an identifier for the C57 sequences or I'll just drown in the results.

    Now where did I have that bone?

    Otis
     
  14. omerbasket

    omerbasket Senior Member

    Messages:
    510
    Likes:
    13
    Is it just me or is there no problem at all here?
    Because non of the viruses that Alter, Lo et al. found are identical to a mouse virus. They might be very close - but they are not identical. So, what's the problem? And alter et al. already proved that these viruses are more closely related to polytropic murine leukemia viruses than to endogenous murine leukemia viruses.
     
  15. Otis

    Otis Señor Mumbler

    Messages:
    1,116
    Likes:
    116
    USA
    I don't think anyone's saying there's a problem. This is an examination about what we know about the Lo/Alter sequences, not some kind of smear campaign.

    But at this point they're (very) partial sequences and (speaking for myself) have been trying to see what we know fits in with known sequences. And I'm perfectly willing to admit I'm stumbling around on the dark here, but I've enjoyed playing with BLAST and learning a thing or two.

    But I'll defer to Coffin when he says "As yet, we can really only refer to one virus, and thats XMRV, and thats only one of the sequences that we know is in a replicating viruses. And I think its very important to keep that in mind. We dont know what if any virus the sequences that match in to PMV or MPMV clades other than XMRV, what their biology is."
     
  16. George

    George waitin' fer rabbits

    Messages:
    851
    Likes:
    80
    South Texas
    Hi Otis
    Sorry I've been so brain dead lately. It comes and goes and when it's bad it's very bad. Just got up from about a 3 day nap (grins) I currently have 3 working brain cells and thought I'd try to use them.

    The whole BLAST thing sounds really cool. I'm guessing here that what is needed to do BLAST is the gene sequences so that you can run two against each other?????

    http://www.ncbi.nlm.nih.gov/genomes/GenomesHome.cgi?taxid=10239

    Looks like a primer on the BLAST

    http://www.ncbi.nlm.nih.gov/sites/entrez?db=nucleotide

    This looks like where to get the nucleotide sequences like the ones from the Alter Paper
    When I typed in XMRV I got 638 hit's some with patents????? and the complete genomes of VP 62/42/35 (these are at the very end of the 638 listed)

    http://www.ncbi.nlm.nih.gov/nuccore/EF185282.1
    This link is the entire nucleotide sequence, I think you can use the Accession number in the BLAST????? and it loads the sequence?????

    On the Friend Murine Leukemia virus there is a whole genome on the FB29 it's link is here http://www.ncbi.nlm.nih.gov/nuccore/Z11128.1 there does seem to be a lot of variants of this virus, A LOT of variants.

    The Moloney MLV (which is classed as a PMLV) is here http://www.ncbi.nlm.nih.gov/nuccore/J02255.1

    Now what I can't figure out is how to find out if a MLV sequence comes from a particular mouse strain. Like what Vince was talking about with the, and here's another fuzzy, either an MLV sequence from Alter/Lo paper was a match for the 129 mouse strain or XMRV is a match for the 129 mouse strain. But where does one find MLV's listed with reference to specific lab mice???? (head scratch)

    I did find this information on the Jackson Laboratories site http://jaxmice.jax.org/strain/004178.html about the 129 strain. (if you click on the tabs it get's into some facinating reading if you have time to read through it, grins) and that lead to this page http://jaxmice.jax.org/protocolsdb/...P2_MASTER_PROTOCOL_ID,P2_JRS_CODE:5200,004178 which list's phenotyping, primers and other information for laboratory use but I still don't know how to match it all up and come to a reasonable piece of information that we can use. ?????

    You probably have all the above information and are happily BLASTING away. (grins) but just in case I could add anything I though I toss this stuff out there. Who knows maybe the links will come in handy down the road. (big grins)
     
  17. Recovery Soon

    Recovery Soon Senior Member

    Messages:
    380
    Likes:
    38
    Dr. Alter and Dr. Enlander

    I just donated blood at Dr. Enlander's office for an XMRV study with Dr. Alter.

    This partnership may be well known, but I didn't know about it. Nor do I have anymore details.

    But they took about 6 vials- I can find out more in 2 weeks.
     
  18. urbantravels

    urbantravels disjecta membra

    Messages:
    1,333
    Likes:
    517
    Los Angeles, CA
    Exciting! Will they give you a result from the test?
     
  19. Recovery Soon

    Recovery Soon Senior Member

    Messages:
    380
    Likes:
    38
    I haven't spoken with him directly about it yet- I will in two weeks- but one of the support staff said if they're available she'll get them for me.

    And I believe it will be tested by 5 different labs.
     
  20. Otis

    Otis Señor Mumbler

    Messages:
    1,116
    Likes:
    116
    USA
    George,

    I'm not entirely sure how one patents a sequence but I'm only going to consent to a study if any and all patents for the little buggers messing me up are ALL MINE. ;)

    You ran across some good stuff. I had found VP62, which is very (99+%) similar to the 2 complete Lombardi sequences I found. FYI - you can further qualify a search by entering 'XMRV Lombardi' and you get all the XMRV sequences associated with the 1960's Green Packers - or the Science study. Whoever drew more blood. :Retro smile: Since there are six sequences it must be the Science Study.

    The genomes list was very interesting and merits more poking around.

    Basically you can compare known sequences against each other, or against the genome DBs, the mouse on being of interest here. I have a lot to learn about all the options, graphical views, etc. - although it's easy to create a phytogenetic tree given a list of sequences.

    For anyone who wants to play with this here's easy example for any fellow geeks out there:
    - If you follow this link
    - Then paste GQ483509.1 in the box below the text "Enter accession number, gi, or FASTA sequence"
    - Make sure the check box next to "Align two or more sequences" is checked
    - Then paste GQ483508.1 in the (second) box below the text "Enter accession number, gi, or FASTA sequence"
    - Then go to the bottom of the page and click the big 'BLAST' button

    If my example was any good you've just compared two partial sequences from the Science paper.

    At the bottom is the interesting part, the results depicting the 'alignment'. It looks (something) like the following:

    >gb|GQ483509.1| Xenotropic MuLV-related virus isolate WPI-1138 putative polyprotein
    gene, partial cds
    Length=366

    Score = 652 bits (353), Expect = 0.0
    Identities = 359/362 (99%), Gaps = 0/362 (0%)
    Strand=Plus/Plus

    Query 1 TTGCAGCACTGGGGAGATGTCCAGCGCATTGCATCCAACCAGTCTGTGGATGTCAAGAAG 60

    Sbjct 1 TTGCAGCACTGGGGAGATGTCCAGCGCATTGCATCCAACCAGTCTGTGGATGTCAAGAAG 60

    Query 61 AGGCGCTGGGTTACCTTCTGTTCCGCCGAATGGCCAACTTTCAATGTAGGATGGCCTCAG 120

    Sbjct 61 AGGCGCTGGGTTACCTTCTGTTCCGCCGAATGGCCAACTTTCAATGTAGGATGGCCTCAG 120

    Query 121 GATGGTACTTTTAATTTAGGTGTTATCTCTCAGGTCAAGTCTAGAGTGTTTTGTCCTGGT 180

    Sbjct 121 GATGGTACTTTTAATTTAGGTGTTATCTCTCAGGTCAAGTCTAGAGTGTTTTGTCCTGGT 180

    Query 181 CCCCACGGACACCCGGATCAGGTCCCATATATCGTCACCTGGGAGGCACTTGCCTATGAC 240

    Sbjct 181 CCCCACGGACACCCGGATCAGGTCCCATATATCGTCACCTGGGAGGCACTTGCCTATGAC 240

    Query 241 CCCCCTCCGTGGGTCAAACCGTTTGTCTCTCCTAAACCCCCTCCTTTACCGACAGCTCCC 300

    Sbjct 241 CCCCCTCCGTGGGTCAAACCGTTTGTCTCTCCTAAACCCCCTCCTTTACCGACAGCTCCC 300

    Query 301 GTCCTCCCGCCCGGTCCTTCTGCGCAACCTCCGTCCCGATCTGCCCAATACACTGCCCTT 360

    Sbjct 301 GTCCTCCCGCCCGGTCCTTCTGCGCAACCTCCGTCCCGATCTGCCCTTTACCCTGCCCTT 360

    Query 361 AC 362

    Sbjct 361 AC 362


    A few notes. There are vertical bars between the top (subject) and bottom (Query) rows, but the spacing came out all wrong even using courier font. :confused:

    For places where a pair isn't the same there is no vertical bar. There are three such cases in this example, near the bottom.

    I bolded a couple of things.
    - Identities are the number of total matches over the total in the comparison.
    - Gaps are places where there is a gap (or insertion) in the compared sequences. For these two there are no gaps.

    Here are the two complete Lombardi sequences IDs which will prove to be a more interesting example.
    GQ497344.1
    GQ497343.1

    Happy blasting.
    Otis
     

See more popular forum discussions.

Share This Page