F.
CSM334 Class 1 SEM
Bioinformatics Databases and Key Online Resource
Ganbat Ganbaatar
2024-2025-A
Overview:
The purpose of this lab session is to introduce a range of bioinformatics databases and associated services
available online whilst investigating the molecular basis of a common human disease.
Sections 1 and 2 deal with querying and searching GenBank, GENE and OMIM databases at NCBI.
Sections 3 and 4 provide exposure to EBI resources for comparing proteins and visualizing protein
structures. Finally, section 5 provides an opportunity to explore these and other databases further with
additional examples.
Side-note: The Web is a dynamic environment, where information is constantly added and
removed. Servers “go down”, links change without warning, etc. This can lead to “broken”
links and results not being returned from services. Don’t give up - give it a second go and try
a search engine using terms related to the page you are trying to access.
Section 1
The following transcript was found to be abundant in a human patient’s blood sample.
>example1
ATGGTGCATCTGACTCCTGTGGAGAAGTCTGCCGTTACTGCCCTGTGGGGCAAGGTGAACGTGGATGAAG
TTGGTGGTGAGGCCCTGGGCAGGCTGCTGGTGGTCTACCCTTGGACCCAGAGGTTCTTTGAGTCCTTTGG
GGATCTGTCCACTCCTGATGCAGTTATGGGCAACCCTAAGGTGAAGGCTCATGGCAAGAAAGTGCTCGGT
GCCTTTAGTGATGGCCTGGCTCACCTGGACAACCTCAAGGGCACCTTTGCCACACTGAGTGAGCTGCACT
GTGACAAGCTGCACGTGGATCCTGAGAACTTCAGGCTCCTGGGCAACGTGCTGGTCTGTGTGCTGGCCCA
TCACTTTGGCAAAGAATTCACCCCACCAGTGCAGGCTGCCTATCAGAAAGTGGTGGCTGGTGTGGCTAAT
GCCCTGGCCCACAAGTATCACTAAGCTCGCTTTCTTGCTGTCCAATTT
The only information you are given is the above sequence so you must begin your investigation with a
sequence search - for this example we will use NCBI’s BLAST service at: http://blast. ncbi.nlm.nih.gov/
Side-note: There are several different “basic BLAST” programs available at NCBI
(including nucleotide BLAST (BLASTn), protein BLAST (BLASTp), and translated
nucleotide nucleotide (BLASTx). We will explore using all of these in the coming sections.
1
Q1 Which BLAST program should we use in this case?
Ď Tip
What type of sequence are you provided with?
Searching against the “Nucleotide collection” (NR database) that includes GenBank is a good place to
start your investigation of this sequence.
Q2What are the names and accession numbers of the top four hits from your BLAST
search?
Q3.What are the percent identities, coverage and E-values for these top few hits?
Ď Tip
These three metrics (E-value, coverage and identities) are the most important for
us to consider at this stage. I suggest you have a discussion with your neighbor and
teacher to make sure you have a firm grasp of these concepts as you will need them
later in your project.
To investigate these results further click on the Alignments section (tab) of your BLAST result page
(Figure 1). This will give you more details on matched nucleotides and important links to “Related
information” about a given “subject sequence”.
Side-note: In BLAST termonology we talk about query sequence and subject sequence. The
query being the input sequence you searched with and the subject being the identified hit
sequence from the database.
2
Figure 1: The BLAST Alignemnts tab contains more detailed information about your results
Q4.How many identical and non identical nucleotides are there in your top hit compared
to your last reported hit?
Ď Tip
Scroll down to the end of the Alignments page to lower ranked hits?
From the results of your BLAST search you can link to the GENE entry for one of your top hits. This
link is located under the “Related Information” heading at the right hand side of each displayed alignment
on the “Alignments” tab (Figure 1).
Q5. What is the “Official Symbol” and “Official Full Name” for this gene?
3
Q6. What chromosome is this gene located on?
Note that there is a rather basic schematic diagram of neighboring genes and their orientations in the
“Genomic context” section that you can use for answering the next question (Figure 2).
Our HBB gene is in maroon. All other gene arrows can be hovered over for full names and clicked on to
link to that specific GENE page to find out more. We will explore more full featured genome browsers at
ENSEMBE and UCSC in an upcoming lab but basically it is the same idea here in a more simplified
form.
Q7. What are the names of neighboring genes on this chromosome?
Q8. How many exons and introns are annotated for this gene?
4
Figure 2: NCBI GENEs Genomic context section indicating location, structure and neighboring genes
Q9What is the function of the encoded protein?
Ď Tip
In addition to reading the abstract like text on a given GENE entry I encourage
you to explore the linked
Gene “ Ontology” information and discuss with your
neighbor and teacher the advantages you think such controlled terms might have
over free text?
5
Q10. Does the protein have a role in human disease(s)? If so what diseases?
Ď Tip
Scroll down to the “Phenotypes” section of the GENE entry page and also explore
the link to the
OMIM database
Section 2
By now you should be aware that there are a number of human diseases linked to particular variants of
the beta-globing gene. In this case our example sequence corresponds to human sickle cell beta-globin
mRNA with this disease resulting from a point mutation in the beta globin gene. In the following section,
you will compare sickle cell and normal beta globin sequences to reveal the nature of the sickle cell
mutation at the protein level.
To do this you need to find at least one sequence representing the normal beta globin gene.
Open a new window and visit the NCBI home page (http://www.ncbi.nlm.nih.gov) and select
“Nucleotide” from the drop menu associated with the top search box. Then enter the search term: HBB
(Figure 3).
Figure 3: The main Google like search tool for searching accross all NCBI databases is called
Entrez. Note that productive use often requires the use of addational “filters” as we will
explore later.
Note that lots of often irrelevant results are returned so lets apply some “Filters” (available by clicking in
the left-hand sidebar) to focus on RefSeq entries (under “Source databases”) for Homo sapiens (under
“Results by taxon” on the right-hand sidebar in this later case).
Side-note: Boolean operators (NOT, AND, OR) as well as fielded queries
(i.e. “HBB[Gene Name] AND Human[Organism]” ) can be used directly in
ENTREZ searches to filter results for more efficient searching.
6
Remember that we are after mRNA so we can compare to the mRNA sequence from section 1 above.
Q11. What is the ACCESSION number of the “Homo sapiens hemoglobin, beta (HBB),
mRNA” entry?
Select “Homo sapiens hemoglobin, beta (HBB), mRNA” from the results and scroll down to the
“FEATURES” section to answer the following. You can also find some of this same information from
selecting the “GRAPHICS” display format and, for example, placing your mouse over the first exon (see
Figure 4).
Figure 4: The GRAPHICS view of a GenBank entry can be more user friendly than the traditional text of
the coresponding GenBank format display.
Q12. What are the numbers of the first and last base positions of exon 1 of this entry?
7
Q13. What are the numbers of the first and last base positions of the CDS?
Ď Tip
CDS or “coding sequence” refers to the portion of a genomic DNA sequence that is
translated, from the start codon to the stop codon. Successful translation of a CDS
results in the synthesis of a protein.
Section 3
Here we will compare the retrieved sequences by creating a sequence alignment. This will make the
difference between the two sequences easy to spot.
To generate the alignment, we will use MUSCLE available on the EBI website at: http:
//www.ebi.ac.uk/Tools/msa/muscle/
Select the FASTA display for the “Homo sapiens hemoglobin, beta (HBB), mRNA” (NM_000518) entry
from section 2.
Now copy-and-paste this FASTA format sequence and also the example1 sequence from section 1 into
the input box of the MUSCLE page. Then click the submit button (see red circle in Figure 5).
Side-note: If your alignment is incomplete, please wait until the page refreshes. If the job
appears to be in an undefined state try clicking refresh until a result is returned.
The two sequences should now be aligned. Where the aligned sequences are identical, an * is placed
under the alignment. Examine the results and note that your sequences are nearly identical. However,
being much shorter, the sickle cell sequence has many padding gap characters (-----) to bring equivalent
regions into the correct register (Figure 6).
When inspecting alignments (especially those with lots of sequences) it can be helpful to use a graphical
user interface (or GUI) to display colored, interactive and scrollable versions of your alignment. One
such GUI is the seaview program.
From your muscle results web page click Download Alignment File (red highlight in Figure 6). Note
that if a download does not automatically begin then you may need to save the resulting plain text page
from your web browser via File> Save As…
8
Figure 5: To use the EBI MUSCLE server you must paste multiple FASTA format sequences that include
their ID lines.
Figure 6: Alignment of patient and reference HBB sequence
9
Next download seaview for your computer from: http://doua.prabi.fr/software/seaview
Once downloaded open seaview by double clicking on it’s icon (most likely in your Downloads folder)
and then select File > Open > and select your muscle alignment results. A colored version of your
alignment should now be displayed (Figure 7).
Figure 7: Programs like SEAVIEW are most useful when you have large many sequence alignments
See if you can now use seaview to answer the following 3 questions:
Q14. How many gap characters (-) are added to the beginning of the sickle cell betaglobin sequence in
order to align it with the beta globin sequence? How might you have
guessed this number from information you read in the GenBank annotation?
Ď Tip
See section 2, Q13.
Q15. Ignoring ambiguity codes (Y and N), what are the differences between the two
sequences?
Ď Tip
There may be more than one
10
Q16. Which codon position from the start of the sickle cell sequence would this difference
affect? What amino acid would the different codons encode in the two sequences?
Ď Tip
Use the codon table below to help8).(Figure
Figure 8: Codon table for use in Q16
11
Section 4
In this section we will retrieve and visualize the 3D protein structure of sickle cell haemoglobin. The aim
here is to ascertain how the Glu6 -> Val6 mutation might cause the mutant molecules to oligomerise into
fibers, hence deforming erythrocytes. This will require you to examine the structural context of the
mutation in the beta globin chains.
We could find sickle cell haemoglobin structures via a text search of main PDB website @
http://www.rcsb.org/. However, as we know the nucleotide sequence from our previous work, lets use
BLASTx to search the PDB database from the NCBI site.
To do this visit http://blast.ncbi.nlm.nih.gov/ select the appropriate BLAST program and make sure the
database you are searching against is set to “Protein Data Bank (pdb)”.
Note the accession numbers and alignment statistics for the top few hits.
Q17. Are there any PDB structures with100% identity to your example1 query sequence?
Give the PDB codes for these entries.
Ď Tip
Note that this might not be the top listed hit
To further examine these structures we will jump over to the main PDB database as it has more
annotation data and more full featured 3D molecular viewers than NCBI.
Visit http://www.rcsb.org/ and use the 4 character PDB accession code you found previously in your
BLASTx search to pull up each PDB entry you listed in Q17.
From scrolling through this entry you can find out information about the “Experimental Data” (such as
the resolution of the structure and quality of the data collected), “Literature” links (i.e. associated
publications), “Macromolecules” (i.e. protein chains present) and “Small Molecules” (i.e. any ligands or
co-factors that might be present).
From the “Macromolecules” section notice that the hemoglobin structure is composed of multiple alpha
and beta globin molecules corresponding to gene names HBA and HBB.
12
Q18. Which four chain identifiers in the 1HBS structure represent beta globin?
At the top of the PDB entry page click the 3D View tab to pull up an interactive 3D structure view
(Figure 9).
Figure 9: Using the molecular viewer at the PDB database
Under the “Structure” section of the right side control panel change the “Type” of view from
“Assembly” to “Model” (see second red rectangle in Figure 9). This will now display the model observed
in the asymmetric unit of the crystal (i.e. the packing of chains observed in the actual experiment) rather
than the simplified default biological assembly view of the minimal functional form.
Side-note: This is the relatively new Mol* 3D viewer. This viewer also features
at a number of other major databases including UniProt. The user interface is still somewhat
clunky and limited when compared to stand-alone software like PyMol, Chimera or VMD.
However, these sand-alone viewes require consideral download and installion time so we
13
will stick with Mol* for this lab. Essentially there are 3 major components to the viewer
(Figure 10):
Figure 10: The three major components of the Mol* viewer include the Sequence panel, 3Dcanvas and
Control panel.
Notice by rotating the molecule in the 3D Canvas that there are now two hemoglobin molecules displayed
rather than the previous single molecule. Notice that each is comprised of four distinctly colored chains
with two alpha and two beta chains in each.
To highlight our beta globin amino acid of interest toggle the sequence display to the beta chain (Figure
11) and specifically chain H (Figure 12) in the top “Sequence panel”.
Figure 11: Display the sequence of beta globin chains in the PDB entry
Find and click on amino acid V 6 in the sequence view to highlight this amino acid in the 3D
Figure 12: Focus on beta globin chain H
14
view (Figure 13). Notice that the right side control panel now lists VAL 6 in several locations (purple
highlight in Figure 13).
Try alternating the zoom level by clicking on these side panel entries (purple in Figure 13) to get a better
feel for the location of our Valine amino acid in the overall structure.
Figure 13: Highlighting Val 6 in the sequence and structure of 1HBS chain H
Using the “Control panel” we can add a new “Representation” to more clearly display the mutated Valine
residue. First click on the 3 dots “action” menu icon beside the [Focus] Target component of the side
panel. This should correspond to our Valine amino acid (Figure 14). Then click “Add Representation” and
select spacefill from the dropdown list of options.
This will result in all atoms of our entry being displayed as so called “ sapcefill spheres” with different
atom types in different colors (e.g. oxygens in red, carbons in gray etc.)
Play around with the settings from the spacefill menu and others in this section until you have a
reasonable feel for how the program works. Can you clearly see our mutated residue position?
15
Figure 14: Adding a spacefill represention to our Val 6 residue
Try zooming (via scrolling up and down) and rotating (via clicking and moving your mouse). You can
always “reset” the view by clicking the reset like circular arows icon. Also experiment with different
settings and views. You can render a
Q19. What do you notice about the location of the Val6 residue in chain H of the 2HBS
structure in relation to porphyrin?
Ď Tip
See Figure below where I have used a white “spacefill” representation for Val6 and
changed the viewer background to black.
In general black backgrounds can help with interactive visualization but should not be used for
publication quality figures. You can download a high resolution image with transparent background of
your final representation using the iris like screenshot icon.
Side note: Some folks have reported issues using the Mol* viewer with older versions of the
Edge, IE and Chrome browsers. The workaround is to use a different web browser. If, the
structure is still not displayed correctly for you, download
16
Figure 15: In this representation, the background has been changed to black and E6V (i.e. Val6) mutation
is highlighted in white spacefill representation and the porphyrin prosthetic group in ball and
stick representation).
its coordinates from the PDB database at: http://www.rcsb.org/ and ask for assistance.
Q20: What one part of this exercise or associated lecture material is still confusing? If appropriate
please also indicate the question number from this document and answer the question in the
following anonymous form: Mudy_Point_Assesment_Form Your comments will let us know
which material needs to be further clarified and will help us gain stronger control of the material in
this course. Thank you!
Discussion
The original paper discussing the 1HBS and 2HBS crystal structures is available online:
http://www.sciencedirect.com/science/article/pii/S0022283697912535
In this article, Figure 3 demonstrates how the Glu6->Val6 mutation could result in the characteristic
“sickle” phenotype. The charged Glu6 mutating to Val6 creates a superficial hydrophobic patch on one
17
HbS molecule that interacts with hydrophobic surface residues of another. The molecules thus
polymerize, creating extended fibers that distort the shape of the red blood cell.
Assessment of the disparate biochemical properties of normal and sickle haemgolobin, together with
microscopy studies showing long crystal fibres inside sickle cells, led Linus Pauling (1949) to (correctly)
predict the morphological effects of these changes. The abnormal sickle form causes the cells to clump
together, hampering their passage through blood vessels, depriving tissues of oxygen. See this YouTube
video for an illustration: http://www.youtube.com/ watch?v=Qd0HrY2NlwY
The sickled blood cells have a short lifetime and cannot be replaced fast enough, leading to chronic
anaemia. Sickle cell anemia was one of the first diseases to be linked to a defect at the molecular level,
providing a clear demonstration that a single base mutation can change a single amino acid, which in turn
can result in a defective protein.
Bluebird Bio as well as Vertex Pharmaceuticals and CRISPR Therapeutics have ongoing sickle-cell gene
therapy trials. As noted in this excellent recent easy in the New Yorker: “There is something of a
paradox in the fact that patients with sickle-cell disease — a population that has faced extraordinary
levels of bias, neglect, and marginalization - may be among the first to have their illnesses transformed by
the most cutting-edge of medical technologies”.
Appendix
>gi|179408|gb|M25079.1|HUMBETGLA Human sickle cell beta-globin mRNA
ATGGTNCAYYTNACNCCNGTGGAGAAGTCYGCYGTNACNGCNCT
NTGGGGYAAGGTNAAYGTGGATGAAGYYGGYGGYGAGGCCCTGG
GCAGNCTGCTNGTGGTCTACCCTTGGACCCAGAGGTTCTTNGAN
TCNTTYGGGGATCTGNNNACNCCNGANGCAGTTATGGGCAACCC
TAAGGTGAAGGCTCATGGCAAGAAAGTGCTCGGTGCCTTTAGTG
ATGGCCTGGCTCACCTGGACAACCTCAAGGGCACCTTTGCCACA
CTGAGTGAGCTGCACTGTGACAAGCTNCAYGTGGATCCTGAGAA
CTTCAGGCTNCTNGGCAACGTGYTNGTCTGYGTGCTGGCCCATC
ACTTTGGCAAAGAATTCACCCCACCAGTGCANGCNGCCTATCAG
AAAGTGGTNGCTGGTGTNGCTAATGCCCTGGCCCACAAGTATCA
CTAAGCTNGCYTTYTTGYTGTCCAATTT
>gi|28302128|ref|NM_000518.4| Homo sapiens hemoglobin, beta (HBB),
mRNA
ACATTTGCTTCTGACACAACTGTGTTCACTAGCAACCTCAAACA
GACACCATGGTGCATCTGACTCCTGAGGAGAAGTCTGCCGTTAC
TGCCCTGTGGGGCAAGGTGAACGTGGATGAAGTTGGTGGTGAGG
CCCTGGGCAGGCTGCTGGTGGTCTACCCTTGGACCCAGAGGTTC
TTTGAGTCCTTTGGGGATCTGTCCACTCCTGATGCTGTTATGGG
CAACCCTAAGGTGAAGGCTCATGGCAAGAAAGTGCTCGGTGCCT
TTAGTGATGGCCTGGCTCACCTGGACAACCTCAAGGGCACCTTT
GCCACACTGAGTGAGCTGCACTGTGACAAGCTGCACGTGGATCC
TGAGAACTTCAGGCTCCTGGGCAACGTGCTGGTCTGTGTGCTGG
18
CCCATCACTTTGGCAAAGAATTCACCCCACCAGTGCAGGCTGCC
TATCAGAAAGTGGTGGCTGGTGTGGCTAATGCCCTGGCCCACAA
GTATCACTAAGCTCGCTTTCTTGCTGTCCAATTTCTATTAAAGG
TTCCTTTGTTCCCTAAGTCCAACTACTAAACTGGGGGATATTAT
GAAGGGCCTTGAGCATCTGGATTCTGCCTAATAAAAAACATTTA
TTTTCATTGC
http://www.rcsb.org/pdb/files/2hbs.pdb
The mutation causing sickle cell anemia is a single nucleotide substitution (A to T) in the codon for
amino acid 6. The change converts a glutamic acid codon (GAG) to a valine codon (GTG). Changing a
hydrophilic amino acid to a hydrophobic one, see
http://themedicalbiochemistrypage.org/sicklecellanemia.php
Note there is also a T -> A difference at position 162 (162/3 => codon 54 GCT -> GCA).
This is in the third position of the codon and hence does not change the corresponding aminoacid.
19
