BLASTing Flu Viruses

[swimmy]

I wanted to try the BLASTing Flu Viruses project. However, the BLAST at ISD doesn't work. Would the BLAST at NCBI be the same? If not, please tell me what is different.

The vaccines are on the ISD page link given in the project, right?

If those are the vaccines, then where would the raw sequences for the actual flu be? How could I find them?

Should I compare the Hemagluttinin protein of the vaccines and viruses or the Neuraminidase protein?

Are the vaccines ones that are currently being used?

Could I compare how similar all the vaccines are to swine flu?

I also don't really get the results that appear after trying the experimental procedure with NCBI's BLAST.

Should I just enter the DNA sequence of the vaccine, or Align two or more sequences? If I do just the DNA sequence of the vaccine, then it says "No significant similarities found" and there are no results.

NOTE: I found the Influenza Research Database, which has sequences as well as BLAST.
Here is the site:
http://www.biohealthbase.org/GSearch/blast.do?method=ShowCleanLaunchPage&decorator=Influenza

Is this the same as ISD? There was a link that led here from ISD home page.

[davidkallman]

Hi swimmy,

Wow! I'm impressed by the depth of knowledge in your post, particularly in Grades 6-8. I don't profess to know answers to your questions. Here's a method that may help getting answers. If you type "blast" into the "Search" box at the upper right of the sciencebuddies main page, you get 57 hits before yours, in the range of 8/2/2005 to 1/4/2009.

Some of these, and the pages they reference, may address the topics in your post. Do they help?

By the way, of your terms, ISD, NCBI, Hemagluttinin and Neuraminidase, only NCBI has references on sciencebuddies, 63 (more than blast!) hits besides yours. Is it a case where there's smoke, there's fire? So, ISD, Hemagluttinin and Neuraminidase may need more research.

Does this help?

[swimmy]

Some of the pages do help, but sometimes some of the people posted in many different forums, or they have different information about the things I am wondering about.

Here are questions I still need answers to:

The page with the sequences of the vaccines does not work. http://www.flu.lanl.gov/vaccine
Could I use sequences of vaccines from http://www.biohealthbase.org/GSearch/vaccineRecommend.do?decorator=influenza instead?

Where would the raw sequences for the actual flu be? How could I find them? Are they on the NCBI page see http://www.ncbi.nlm.nih.gov/genomes/FLU/SwineFlu.html

Should I compare the Hemagluttinin protein of the vaccines and viruses or the Neuraminidase protein?

Are the vaccines (http://www.flu.lanl.gov/vaccine or http://www.biohealthbase.org/GSearch/vaccineRecommend.do?decorator=influenza (if it can be used) ones that are currently being used?

Could I compare how similar all the vaccines are to swine flu? To find out if they are effective for the swine flu outbreak?

I also don't really get the results that appear after trying the experimental procedure with NCBI's BLAST. What percentage should be matched for the vaccine to be effective?

Should I just enter the DNA sequence of the vaccine, or Align two or more sequences? If I do just the DNA sequence of the vaccine, then it says "No significant similarities found" and there are no results.

I also don't get the "database" comparing to flu sequences. Which one should I use if any?

Also, what vaccine should I choose?

I know these are still most of the questions I asked before, but after searching "BLAST" on the search, then I still couldn't find the information to answer these questions or there was ambiguity in the answers.

[MelissaB]

Hi,

First of all, yes, you can use vaccine sequences from the site you found.

I find the CDC website (http://www.cdc.gov/flu/weekly/fluactivity.htm) to be most helpful for figuring out which flu strains went around. For example, if you look at the summary of the 2007-2008 season, it will tell you what sequence most strains were 'antigenically similar to'. You can then use the name of this strain to look up the sequence in BLAST. For example, in 2007-2008, 66% of type A infections were similar to A/Solomon Islands/3/2006.

It is up to you whether you want to compare the H protein or the N protein.

As far as I know, the WHO recommendations are indeed the vaccines that were used (at least in North America). For earlier years, the CDC pages I pointed you to above list the sequences in the vaccines.

On this site: http://www.ncbi.nlm.nih.gov/genomes/FLU/FLU.html you can find 'sequences from the pandemic (2009) viruses', which should give you some swine flu sequences to work with. So yes, you could see whether or not the vaccine sequences are close to the swine flu sequences.

As for effectiveness, you can usually get a good estimate by looking at the proportion of samples that are antigenically similar to the proteins in the vaccine. But generally, the closer the match, the more likely the vaccine will work.

It is up to you which vaccine you use! You might consider comparing the similarities between this year's normal flu vaccine, the swine flu vaccine, and the swine flu--it sounds as if you're interested in whether or not the 'normal' vaccine will work, right?

Definitely align two different sequences with BLAST--the vaccine, and the flu sample you're interested in.

Take a good look through the CDC website and the information it has, then post back if you have more questions.

[swimmy]

Thanks, this really helps.
swimmy

[amyc]

Hi Swimmy - I have been in contact with the team of scientists here at Science Buddies and the author of the project you have been working on. Based on the questions you've raised, we are editing the project some to make parts of it clearer. I have some answers to pass along to you from the project author. I hope these help. I'm copying your question in here along with his answer:

1. The page with the sequences of the vaccines does not work. http://www.flu.lanl.gov/vaccine - OUT OF COMMISSION
Could I use sequences of vaccines from http://www.biohealthbase.org/GSearch/va ... =influenza instead?

Answer: Yes. But it might be easier to use the NBCI site: http://www.ncbi.nlm.nih.gov/Genbank/

2. Where would the raw sequences for the actual flu be? How could I find them? Are they on the NCBI page?

Answer: See http://www.ncbi.nlm.nih.gov/genomes/FLU/SwineFlu.html

3. Should I compare the Hemagluttinin protein of the vaccines and viruses or the Neuraminidase protein?

Answer: Use the hemagluttinin (HA) protein sequence. This is the protein that is assayed by the hemaglutination assay that is used to test how well the vaccine is working. And protein sequence is more informative that dna since dna accumulates silent mutations that complicate interpretation of the blast results.

4. Could I compare how similar all the vaccines are to swine flu? To find out if they are effective for the swine flu outbreak?

Answer: Yes. Go to the NCBI Flu page for the sequence: See http://www.ncbi.nlm.nih.gov/genomes/FLU/SwineFlu.html

5. I also don't really get the results that appear after trying the experimental procedure with NCBI's BLAST. What percentage should be matched for the vaccine to be effective?

Answer: It is hard to say up front what percentage should be matched. I suggest that you formulate a new question that the BLAST output is able to address. I have not tried any of the projects below, so they may not be good ways to proceed, but they will give you an idea of the kind of project I have in mind. For example, you could ask

A) What part of the protein is most subject to change? (Might try to form a hypothesis as to why certain parts of the protein are more subject to change). Might want to focus on one variation to make the project manageable. For example, this part of the HA protein has a change from a D to an N.

Query 121 YASLRSLVASSGTLEFNDESFNWTGVTQNGT
YASLRSLVASSGTLEFN+ESFNWTGVTQNGT
Sbjct 121 YASLRSLVASSGTLEFNNESFNWTGVTQNGT

B) How are various strains distributed geographically?

C) Can you identify when (what year) a certain variant appeared and how it spread?

6. Should I just enter the DNA sequence of the vaccine, or Align two or more sequences? If I do just the DNA sequence of the vaccine, then it says "No significant similarities found" and there are no results.

Answer: Use protein sequence.

7. I also don't get the "database" comparing to flu sequences. Which one should I use if any?

Answer: Use the protein database

8. Also, what vaccine should I choose? I would pick a year for which there is good molecular data, and not too recent. Say 2004.

Answer: It is up to you!


I hope these help, Swimmy. If you have additional questions, please post here, and I'll follow up with the scientist. I'll also post here and let everyone know when the project is officially updated.

Amy Cowen
Science Buddies

[swimmy]

Thanks, Amy.
Ok, I understand where to find the raw sequences of the actual flu virus. I checked the site (http://www.ncbi.nlm.nih.gov/Genbank/) but I don't really get how to get the flu vaccine sequence still.
I searched "Are normal season flu vaccines effective for swine flu?" on google. There are varied results, so I want to go with one of the ideas that you mentioned on #5. I want to do #5 (A), but I don't really get how to go about doing that.

By protein database, do you mean Protein Data Bank (pdb)? This is the only protein database I could find in blastn.

[amyc]

Hi Swimmy - I have some good news! The Project Idea has been updated with new directions that I hope will help clear up some of the trouble you ran into. The project is here:

http://www.sciencebuddies.org/science-f ... p003.shtml

After looking it over, if you have questions, let us know.

Amy
Science Buddies

[swimmy]

Thanks again, Amy.

CDC has antigenically characterized 2,112 seasonal human influenza viruses [1,189 influenza A (H1), 227 influenza A (H3) and 696 influenza B viruses] collected by U.S. laboratories since October 1, 2008, and 493 2009 influenza A (H1N1) viruses.

All 1,189 seasonal influenza A (H1) viruses are related to the influenza A (H1N1) component of the 2008-09 influenza vaccine (A/Brisbane/59/2007). Two hundred sixteen (95%) of 227 influenza A (H3N2) viruses tested are related to the A (H3N2) vaccine component (A/Brisbane/10/2007) and 11 viruses (5%) tested showed reduced titers with antisera produced against A/Brisbane/10/2007.

"All 493 2009 influenza A (H1N1) viruses are related to the A/California/07/2009 (H1N1)pdm reference virus selected by WHO as a potential candidate for 2009 influenza A (H1N1) vaccine."

~CDC

This is about the H1N1 virus. It says the virus is related to a part of the vaccine for seasonly flu.
Should I use those strains mentioned as the vaccines?
Or should I use the candidate strain for the vaccine?

I tried the method in the revised edition of BLASTIng Flu Viruses. I am successful with the procedure but then it says "Putative conserved domains have been detected, click on the image below for detailed results." What does this mean?

Also, the results are kind of confusing, because I don't see the alignments...
also, there are no lowercase letters. I'm not really sure if what I am seeing is correct, if it is correct, I don't know how to analyze the results.

Here is the blast search:http://blast.ncbi.nlm.nih.gov/Blast.cgi?PAGE=Proteins&PROGRAM=blastp&BLAST_PROGRAMS=blastp&QUERY=ACQ55359.1&PAGE_TYPE=BlastSearch&LINK_LOC=protein&log$=seqview_box_blast

here is the HA protein I used http://www.ncbi.nlm.nih.gov/protein/ACQ55359?ordinalpos=1&itool=EntrezSystem2.PEntrez.Sequence.Sequence_ResultsPanel.Sequence_RVDocSum


Why are the letters not just A, G, C and T?
Is it the amino acids?

What do the plus signs (+) and the spaces mean?

Also how would I be able to see the rate at which antigenic shift occurs? I might use this as an idea...
Would I be able to see the effect of environment on the rate of mutation?

[MelissaB]

Swimmy,

I'll let Amy answer most of your questions, but 'putative conserved domains' just means 'places where the virus doesn't mutate'. These are generally places that have to be conserved in order for the protein to function properly--for example, the parts that anchor in into the virus' membrane, or parts relating to the protein's function.

[swimmy]

In addition to all those questions above,
I have a few questions about writing the report for this project.
If I had the title "Which regions of the influenza virus protein are most subject to change?", what could I write for my hypothesis?
Do I just specify the kind of protein? (That doesn't make much sense to me, because it is just the hemagluttinin protein that I am using)
Would I need to cite where I got the protein sequence?
and also, would i need to cite the websites in the procedure?

I need help really quickly because I need to finish a draft by tuesday, but I'll be busy tomorrow.

[MelissaB]

Hi,

Yes, you should definitely cite the websites you used during your search and tell where you got the protein sequence.

In order to form a hypothesis, you would need to do research on what the various parts of the protein do (are they structural, do they anchor the protein to the membrane, do they interact with other cells, etc.). Then you could guess whether or not they will be subject to change.

[swimmy]

ok, so i have turned in the draft with conclusion saying i didn't know if the results were correct.
so, i still need help on the question from september 4 (last week)
and i would appreciate help for the other question too

[swimmy]

I have done research on different types of proteins in influenza
here are the proteins:
hemagglutinin
neuraminidase
nucleocapsid
nuclear export protein
nonstructural protein 1
Polymerase PA
polymerase PB1
polymerase PB2
matrix protein 1
matrix protein 2

these are all the proteins I could find in GenBank
so should i do blast searches for all of them and then compare how many lowercase letters (i still need help on this)?
if not, what should i do?
and also how should i figure out what region of protein? is it one of the types mentioned above or an actual region?
how would I describe it?

how do i site the sequences?
also, can you please help me understand the paper http://www.biology-direct.com/content/p ... 0-4-18.pdf

[swimmy]

I want to submit this to the school science fair...
I need help identifying the independent/manipulated, dependent/responding, and controlled/constant variables.
What should I use for a graph?

These things are on the application.

Please help on questions posted september 4th and later