Biology 100A ÑBioinformatics Lab

Developed by Alexandra Schnoes, PhD., UCSF, based on a lab exercise by Crima Pogge

  

What is 'bioinformatics'?

Bioinformatics is the utilization of computation for biological investigation and discovery. It is the process by which you unlock the biological world through the use of computers.


What should I read for this lab?

1. Read briefly about viruses in Hillis, p. 210-211. Pay special attention to the virus figure 11.4 on page 211.

2. Skim the Wikipedia page on influenza viruses.

3. Skim the Wikipedia page on the 1918 flu pandemic (aka Spanish Flu Pandemic).

 

Where can I explore bioinformatics?

(!! are recommended sites)

Science Magazine Netwatch !!
(Weekly list of cool science sites. Not always bioinformatics sites per se. A fun way to see how computers are being used in science and to communicate science)

ExPASy Life Sciences Directory !!
(This is the standard web site to go to for bioinformatics programs)

NCBI Portal !!
(The US site for bioinformatics related resources. Maintained by the National Center for Biotechnology Information, the National Library of Medicine and the National Institutes of Health. You can do literature searches, examine genomes, get programs to do sequence analysis and download or search the current protein and nucleotide databases... among other things)

Bioinformatics. Org
(General bioinformatics information)

Bioinfo Online

Bioinformatics Links Directory !!
(Links, links and more links to bioinformatics resources)

Google Bioinformatics Directory

Lab:

The Blame Game

       The greatly feared pandemic flu virus has finally broken out. Millions are sick and thousands have already died. It is almost impossible for the Centers for Disease Control and Prevention (CDC) to keep track of the new cases that are reported each day. Contrary to everyone's expectations, the first reported cases appeared in San Francisco and not in Asia or Eastern Europe. At the same time, the New York Times is reporting that the recently reconstituted and extremely dangerous 1918 influenza virus was seriously mishandled by several scientific laboratories. Apparently, the virus was shipped to scientists at UC San Francisco without using the appropriate shipping procedures. The news report has hypothesized that perhaps the package was damaged en route or potentially mishandled onsite at UCSF. In immediate reaction to the newspaper's report all related parties at UCSF have been arrested for the illegal dissemination of a biological agent to the public. Several of the arrested parties are researchers without US citizenship (but with appropriate visas) and some members of congress are calling for immediate deportation or even reclassification of their status to 'Enemy Combatants' and trying them as terrorists. In other related news, the virus strain from patients in San Francisco has been fully sequenced and, just today, released to the public.

 

Discussion:

 

1. What is this case about?

 

2. What do we need to know?

 

3. Where do we find the information we need?

 

Questions to Answer:
 

Say you're an enthusiastic bioinformatician suffering through a miserable (but not deadly) bout of the new flu. You don't know much about the 1918 flu virus but you're darn unhappy with how sick you feel. How do you figure out whom to blame? Evolution? Or thoughtless, potentially criminal, scientists?

1. The best site for cutting edge research information is PubMed (a searchable site of research article citations).

2. Through your research you determine that the 1918 flu is an RNA virus: Influenza A of type H1N1. Since you have the sequence data from the San Francisco virus you decide to examine some of the sequence to see if it is similar to the 1918 flu sequence. The sequence you have is


>SF strain sequence
TTCATGCAGCAAAAGCAGGTTTAGTACCATGGACAACCAAACAAAGACAATGACTATCACTTTTCTCATC
CTCCTGTTCACAGTAGTGAAAGGGGACCAAATATGTATCGGATACCATGCCAACAATTCCACAGAAAAAG
TCGACACAATCTTGGAGCGAAACGTCACCGTGACTCATGCCAAGGACATTCTTGAGAAAACGCATAATGG
GAAGTTGTGCAGATTAAGCGGGATCCCTCCATTGGAACTGGGGGATTGCAGCATTGCAGGTTGGCTCCTT
GGAAATCCGGAATGTGACCGGCTCTTAAGTGTACCTGAATGGTCCTATATAGTGGAAAAGGAAAACCCGG
CGAATGGTCTGTGCTACCCAGGCAGTTTCAATGATTATGAGGAACTGAAACATCTCCTCACCAGTGTGAC
ACACTTTGAGAAAGTTAAGATTCTGCCCAGAGATCAATGGACCCAGCACACAACAACTGGTGGTTCTCGG
GCCTGTGCAGTATCTGGCAACCCGTCATTCTTCAGAAACATGGTTTGGCTTACAGAGAAGGGGTCAAACT
ACCCAATTGCTAAAAGATCATACAACAACACAAGCGGGAAGCAAATGCTGGTAATTTGGGGGATACATCA
TCCCAATGACGATACGGAACAAAGGACACTGTACCAAAATGTGGGAACATATGTTTCCGTGGGAACATCA
ACACTGAATAAGAGGTCAATCCCTGAAATAGCAACAAGGCCCAAAGTCAATGGACAAGGAGGGAGAATGG
AATTCTCTTGGACTCTATTGGAGACATGGGATGTCATAAATTTTGAGAGCACTGGTAATTTAATTGCACC
AGAATACGGATTCAAAATATCGAAGAGAGGAAGCTCAGGAATTATGAAGACAGAGAAAACGCTTGAAAAT
TGTGAAACCAAATGTCAGACCCCCTTGGGGGCAATAAATACAACACTGCCTTTTCACAACATTCACCCAT
TGACAATAGGTGAGTGCCCCAAATATGTAAAGTCAGATAGATTGGTTAAGGCGACAGGGCTAAGAAATGT
CCCTCAGATTGAATCAAGGGGATTGTTTGGAGCAATAGCCGGGTTTATAGAAGGCGGATGGCAAGGAATG
GTTGATGGCTGGTATGGGTATCATCACAGCAATGATCAAGGATCAGGATATGCAGCAGACAAAGAATCCA
CTCAAAAGGCAATTGATGGGATAACTAACAAAGTAAATTCTGTGATTGAAAAGATGAACACTCAGTTTGA
GGCTGTTGGGAAAGAGTTCAACAATCTAGAGAGAAGACTGGAAAACTTAAATAAAAAGATGGAAGATGGA
TTTCTTGATGTATGGACATATAATGCCGAACTCCTAGTTTTAATGGAAAATGAGAGGACACTTGATTTCC
ATGATTCTAATGTGAAGAATCTGTACGATAAGGTCAGAATGCAGTTGAGAGACAATGCTAAGGAAATAGG
GAATGGATGCTTTGAGTTTTATCATAAATGTGATGATGAATGCATGAATAGTGTCAGGAATGGGACATAT
GATTATCCCAAATATGAAGAAGAGTCCAAGCTGAACAGAAACGAAATCAAAGGAGTGAAATTGAGCAATA
TGGGGGTTTATCAAATACTTGCTATATACGCTACAGTTGCAGGCTCCTTGTCACTGGCAATCATGATAGC
TGGGATTTCTTTCTGGATGTGTTCTAATGGGTCTCTGCAATGCAGAATTTGCATATGACTGTAAGT

A) This sequence is supposed to code for a virus protein. This is an RNA retrotranscribed into a DNA strand sequence. How many reading frames does it have?

B)If it were a true DNA sequence (also given as just one strand but composed of two), how many reading frames would it have?

C) The RNA start codon for translation is 5'AUG3' . In the first reading frame (in the 5'3' direction) locate and label on your worksheet the first corresponding start triplet. (Remember sequences are normally written 5'3').

D) Using this translation table find the first stop codon in the same, first reading frame and mark it on your worksheet.

E)  How many bp (base pairs) long would this gene be?

F)  How many amino acids is this? Is this a reasonable length for a gene?       

G) Label the first start codons in the second and third reading frames in the 5'3' direction.    You don't need to look for stop codons here because...  

H) in the second reading frame there are 1707 bp from the first Met to the stop codon. In the third reading frame there are 54 bp. Which reading frame likely contains the gene?

I) How many amino acids long would the protein product of this gene be?


As you can tell, this can be a time consuming process. Fortunately we have computer programs to help us do this quickly.


3. Go to the Translate Tool on the ExPASy web site. Cut and paste the above sequence into the box. Leave the default information and click on 'translate sequence'. Make sure to only cut and paste the sequence (and not the '>SF strain sequence').

Note: The ExPASy (Expert Protein Analysis System) web server is an extensive and valuable resource put together and maintained by the Swiss Institute of Bioinformatics. A large variety of databases and analysis tools related to proteins are available from them. ExPASy started in 1993 was the very first web server in the field of life sciences.

A) Which reading frame might contain the gene - remember that genes need to have a minimum length? Sequences without a stop codon are called open reading frames and are one of the indicators for coding sequences.

B) What are the first 6 amino acids of the protein?

C) All the amino acids except the start codons (Met) are abbreviated by single letters. What do the single letters of the first 6 amino acids stand for? (If you don't know, google 'amino acid single letter codes')

D) What is the single letter abbreviation of Met?

Click on the frame that contains the gene (i.e. click on the '5'3' Frame...' that matches the frame of the gene). This brings up a page with the sequence of the translation. Click on the Met that starts the gene to bring up a page titled 'Virtual:...' (an informational page about the sequence).


Click on the 'FASTA format' link at the bottom to bring up the FASTA formatted sequence of the protein (note: FASTA format is simply a standard format that people use to write out sequences. Click here to get a full description). Open up a text editor such as Notepad or TextEdit. Cut and paste the full sequence (including the '>virt|...' line) into the editor and save it for later.

 

4. Now that we have both the nucleotide and amino acid sequence of the SF strain viral protein, let's try to find out some functional and evolutionary information about the sequence. Let's compare the sequence we have to other known sequences. We can do this with a BLAST search on the NCBI web site.

Note:  NCBI (National Center for Biotechnology Information) was established in 1988 as a national resource for molecular biology information, NCBI creates public databases, conducts research in computational biology, develops software tools for analyzing genome data, and disseminates biomedical information - all for the better understanding of molecular processes affecting human health and disease

Go to NCBI's BLAST server. Under 'Basic BLAST' click on 'nucleotide blast'. Cut and paste the above nucleotide sequence (scroll back up to the DNA sequence, don't use the protein sequence you just saved) into the search window 'Enter Query Sequence'. Then go to the section labeled 'Database'. Using the drop down list chose the database 'Nucleotide collection (nr/nt)'. Then click 'BLAST' at the very bottom of the page. A new page will come up that will eventually have your results on them (it might take a moment).

A) A new page will come up with a lot of red lines. Each red line corresponds to a sequence significantly similar to the one you "blasted". You can click on one of the lines or scroll down to see its despription. For this exercise, scroll down and look at the top sequence (Influenza A virus (A/mallard/Alberta/202/1996(H2N5)) hemagglutinin precursor (HA) gene, complete cds). To the right of this sequence description, you see some scores (this is from the alignment matrix we discussed), an E-value, and a percentage indicating how much of your sequence is identical to the sequence found.

B) What does an E-value mean? (hint: check out this page). How is this similar to the confidence levels we discussed when doing our statistical analyses?

C) What is the the accession number of the first hit? _________ Click on the accession number to find out the answers to the following questions.

D) Is the top hit a virus sequence?

E) If so, what kind of virus is it (name and type)? (Remember that the 1918 virus is an Influenza A virus of type H1N1).

F) Is the the same type as the 1918 virus?

G) What animal was the sequence isolated from? Where was it isolated?

H) What is the description (name) of the gene? What are name and function of the protein it codes for (remember from the animation we saw or use a search engine)?

5. What we would really like to know is whether or not the SF strain is most closely evolutionarily related to the 1918 flu. So let's spend some time looking at the protein sequence of the gene. Go back to the FASTA amino acid sequence that you saved to Notepad. Go again to the BLAST server and this time click on 'protein blast'. Cut and paste the amino acid FASTA sequence (the '>virt|...' sequence from your text editor) into the search page and click on 'BLAST' and wait for your results.

 

Scroll down to the 'Sequences producing significant alignments' section and look at the top five hits.

A) What are the H and N types of the 5 top hits?

B) What are their accession numbers?

 

Select the top five hits (by clicking on the little box next to the name for each hit). Then click on 'Download' and chose FASTA. This creates a txt file called seqdump.txt with the amino acid sequences of your five alignments. Cut and paste all the sequences into the Notepad document where you have the other amino acid sequence. Make sure to include all of the '>...' lines when you cut and paste.

 

As you can imagine, just staring at lists of sequences is not very illuminating. Let's look at these sequences in a more visual way by aligning the sequences and then looking at the alignment.

Go to the ClustalW sequence alignment site. Now cut and paste all the amino acid sequences from the Notepad file into the query box. Make sure to include the lines that start with '>...'! It turns out that we also have the same protein from the 1918 virus. Add this to the sequences in the query box by cutting and pasting the sequence below into the box after the other sequences (again, make sure to include the '>...' line). You should have 7 sequences in the query box.

>gi|4325018|gb|H1N1_1918 protein
MEARLLVLLCAFAATNADTICIGYHANNSTDTVDTVLEKNVTVTHSVNLLEDSHNGKLCKLKGIAPLQLG
KCNIAGWLLGNPECDLLLTASSWSYIVETSNSENGTCYPGDFIDYEELREQLSSVSSFEKFEIFPKTSSW
PNHETTKGVTAACSYAGASSFYRNLLWLTKKGSSYPKLSKSYVNNKGKEVLVLWGVHHPPTGTDQQSLYQ
NADAYVSVGSSKYNRRFTPEIAARPKVRDQAGRMNYYWTLLEPGDTITFEATGNLIAPWYAFALNRGSGS
GIITSDAPVHDCNTKCQTPHGAINSSLPFQNIHPVTIGECPKYVRSTKLRMATGLRNIPSIQSRGLFGAI
AGFIEGGWTGMIDGWYGYHHQNEQGSGYAADQKSTQNAIDGITNKVNSVIEKMNTQ

 

After you have done this, click on 'Submit' and wait for your results. Scroll down to the alignment and click on 'Show Colors'. The colors help you see when sequences are different from each other.

C) In the first line of sequences, how many times is there an amino acid difference between the 1918 sequence and the SF strain sequence? (remember, the SF sequence will be labeled something like 'virt|...') .

D) Based on this alignment, do you think that the SF strain protein is most like the 1918 protein?

 

6. Counting differences between alignments is tedious, isn't it? A faster way to tell if things are closely related to each other evolutionarily is to look at a phylogenetic tree derived from the sequence alignment.

Note: Phylogenetics is the scientific field that examines the evolutionary relatedness of different species. Molecular phylogenetics uses either nucleotide or amino acid sequences to determine the evolutionary relationships. A phylogenetic tree is the visual representation of the determined evolutionary relationships between the sequences under inspection. An example of a phylogenetic tree is this (from Taubenberger et al., Nature: 437, 889-893, 2005). A well-built phylogenetic tree is a much more reliable indicator of the evolutionary relatedness of two sequences than a simple pair-wise comparison (because phylogenetic trees are build from multiple sequences which allows for a stronger evolutionary signal versus noise).

Let's use the ClustalW site to calculate a phylogenetic tree. Click on "Guide tree". Scroll down to the phylogram and answer the following questions (note, this does not work in Internet explorer 9.0 and you need to have Java properly running).

A) Is the 1918 sequence closest to the SF strain sequence?

 

B) What is the label of the sequence closest to the SF strain sequence?

 

C) Look back at your sequences, what is the name, type, animal and location of that sequence?

 

D) After all this analysis, do you think that the pandemic flu is the 1918 flu virus?

 

E) Should the scientists be prosecuted or let go?

 

F) Why might you still be unsure?

 

7. The reconstitution of the 1918 flu virus has been very controversial. It is one of many ethical debates facing the scientific community today

A) What are the ethical concerns about this research?

B) What are arguments for the virus reconstitution?

C) If the decision were up to you, would you have reconstituted the virus?

D) Given that it has been reconstituted do you trust the scientific community adequately and ethically care for the virus?

E) What regulations/rules would you like to see put in place?