Essential Bioinformatics and Biocomputing LSM2104: Section I Lecture 3: More biological databases, r

1
Essential Bioinformatics and Biocomputing
(LSM2104 Section I) Lecture 3 More
biological databases, retrieval systems and
database searching
2
Biological databases Function and pathways
databases - KEGG

• KEGG (http//www.genome.ad.jp/kegg/kegg2.html)
  database links genetic
• info with cellular functions. It provides keyword
  and pre-calculated sequence
• comparison searches.
• It consists of several interconnected databases
• PATHWAY contains info on metabolic and regulatory
  networks.
• GENES contains information on genes and proteins.
• LIGAND contains information on chemical compounds
  and reactions involved in cellular processes.
• EXPRESSION and BRITE contain micro-array gene
  expression data.
• SSDB helps identify protein coding genes.
• It has an integrated database retrieval system
  DBGET

3
KEGG web-page
4
KEGG Human TCA Pathway
5
Biological databases BINDBiomolecular
Interaction Network Database http//www.bind.ca/

• Stores descriptions of interactions, molecular
  complexes and pathways.
• Provides search tools.
• PreBIND locates literature sources
• Show me a list of all of the papers in PubMed
  that are about my protein of
• interest. Then classify all of these papers and
  tell me which ones are likely
• to contain interaction information.
• Finally, identify all of the other proteins
  mentioned in these papers and
• indicate whether these proteins might interact
  with my protein of interest
• Bader GD, et al. BIND The Biomolecular
  Interaction Network Database.
• Nucleic Acids Res. 2001 29(1)242-5.

6
Biological databases BINDBiomolecular
Interaction Network Database http//www.bind.ca/

• 3.Its Blast searches BIND database for similarity
  to a query sequence.
• BIND is at the forefront of the proteomics
  efforts and is expected
• to grow from the large-scale proteomic data.
• Bader GD, et al. BIND The Biomolecular
  Interaction Network Database.
• Nucleic Acids Res. 2001 29(1)242-5.

7
BIND website
8
BIND Statistics

• Database
  Record Count
• Interaction Database
  11255
• Biomolecular Pathway Database 8
• Molecular Complex Database 851
• Organisms represented
  12
• GI Database
  4651
• DI Database
  0
• Publication Database
  428

9
Protein family/domain databases Sequence
alignmentPfam (http//www.sanger.ac.uk/Software/P
fam/)

• Pfam is a collection of multiple protein sequence
  alignments and statistical models that can be
  used to classify protein families and domains.
• Descriptions of protein domains
• Given an established SWISSPROT sequence, Pfam
  shows pre-computed domain structure of the
  protein.
• Given a completely new protein sequence, Pfam
  computes a domain structure.

10
Pfam Web-site
11
Protein family/domain databases Sequence
patterns PROSITE ( http//ca.expasy.org/prosite/
)

• Protein families and domains. It consists of
  biologically significant sites, patterns and
  profiles that help to reliably identify to which
  known protein family (if any) a new sequence
  belongs.
• It currently contains patterns and profiles
  specific for more than a thousand protein
  families or domains.
• An example of a pattern (motif)
• W-x(9,11)-VFY-FYW-x(6,7)-GSTNE-GSTQCR-
  FYW-x(2)-P

12
Protein sequence motif databases-PROSITE

• A profile is a matrix derived from multiple
  alignments

13
PROSITE web-site
14
PROSITE web-site
15
Biological data retrieval systems Entrez
http//www.ncbi.nlm.nih.gov/Database/index.html

• A retrieval system for searching a number of
  inter-connected databases at the NCBI. It
  provides access to
• PubMed The biomedical literature (Medline)
• Genbank Nucleotide sequence database
• Protein sequence database
• Structure three-dimensional macromolecular
  structures
• Genome complete genome assemblies
• PopSet population study data sets
• OMIM Online Mendelian Inheritance in Man
• Taxonomy organisms in GenBank
• Books online books
• ProbeSet gene expression and microarray datasets
• 3D Domains domains from Entrez Structure
• UniSTS markers and mapping data
• SNP single nucleotide polymorphisms
• CDD conserved domains
• 2. Entrez allows users to perform various
  searches.

16
Entrez Interface
17
Biological data Retrieval systems
SRShttp//srs.ebi.ac.uk/

• SRS is a retrieval system for searching several
  linked databases at the EBI. Similarly to Entrez,
  it provides access to various databases and
  enables various keyword, sequence similarity or
  class searches.

18
SRS interface
19
Biological databases Database searching

• Database searching can be used to answer the
  kinds of question like
• What is the sequence of human IL-10?
• What is the gene coding for human IL-10?
• Is the function of human IL-10 known? What is it?
• Are there any variants of human IL-10?
• Who sequenced this gene?
• What are the differences between IL-10 in human
  and in other species?
• Which species are known to have IL-10?
• Is the structure of IL-10 known?
• What are structural and functional domains of the
  IL-10?
• Are there any motifs in the sequence that explain
  their properties?
• What is an upstream region of IL-10 containing
  transcriptional regulation sites?

20
Biological databases Database searching

• For well studied molecule such as IL-10, we
  expect to extract much of the well-known facts.
• These searches are useful for characterizing
  newly identified sequences
• Notes
• Multiple errors can be found in database
  entries. Some of these errors are introduced with
  the submission of sequences to databases. Some
  errors are due to naming conventions (or lack of
  these). Some errors are due to poor links between
  databases.
• Users should take data extracted from databases
  with care and compare these results with
  information from other databases, journal
  articles, and other sources.

21
Biological databases Keyword searching

• Search DNA and protein databases with keywords
  (10-July-2002)

22
Biological databases Keyword searching

• Notes
• GenPept is protein translation of GenBank. SPTR
  is SWISS_PROT plus protein translation of EMBL
  sequences.
• Different databases contain different, but
  overlapping, sets of entries. The same sequence
  may have entries in different databases
• Some databases have non-redundant sections. For
  example the UniGene System which automatically
  partition GenBank sequences into a non-redundant
  set of gene-oriented clusters.
• For completeness of results usually it is
  necessary to search multiple databases.

23
Biological databases Database coverage

• Example Scorpion KALIOTOXIN 2 (SwissProtP45696)

24
Biological databases Database errors

• Our scorpion study (Srinivasan et al., 2002) also
  revealed numerous errors and missing data in the
  major databases. One of the entries had an error
  in sequence in journal publication, but correct
  sequence in the databases.

25
Biological Databases Sequence Similarity
Searching

• Proteins that have similar sequence often have
  similar structure and similar
• function. If we have only a protein sequence we
  can deduce its
• structural and functional properties by analyzing
  sequences that are similar.
• Sequence Structure Function Relationship
• Similar sequence Similar structure Similar
  function
• Why this relationship?
• Evolution involves sequence variation
• Laws of physics and chemistry defines
  sequence-structure relationship
• Function as defined by molecular interaction
  structure-based

26
Database Searching Cautionary Notes

• Some database matches happen because of chance
  similarities, keywords and sequence similarity
  alike. Distinguishing chance matches from
  biologically significant matches is one of the
  most important issues for effective use of
  biological databases.
• Searching GenBank by sequence similarity tool
  BLAST for short, nearly exact matches, for the
  sequence similarity to the names of the lecturers
  of this module in last semester returned two
  imperfect matches to VLADIMIR and seven perfect
  matches to TINWEE.

27
Database Searching Cautionary Notes
28
Database Searching Cautionary Notes

• If we blindly interpret these results, we would
  erroneously conclude that motif VLADIMIR may have
  some functional importance for structure or
  function of the strawberry vein binding virus,
  and that TINWEE has to do with calciumdependent
  protein kinase in rice and possibly in Legionella
  pneumophila.
• We would avoid conclusions like this by looking
  at the similarity scores. This will be done in
  more detail later in the course, for now it is
  important to know that the lower the expected
  value, the better the match. Anything close or
  greater than 1 should be observed with suspicion.
  However, sometimes matches that are not
  statistically significant, still can have
  biological significance. If we suspect that this
  might be the case, further analysis is necessary.

29
Database Searching Cautionary Notes

• Examples of chance matches virtually any string
  or keyword can show matches to database
  entries. We are interested only in real ones.
• The same search with GenBank. Fortunately we
  have statistical measures that indicate the
  quality of matches. However, sometimes matches
  that have low statistical significance,
  nevertheless have real, biological significance.
  More about that will be taught later in the
  course.

30
Biological databases Concluding remarks

• Biological databases represent an invaluable
  resource in support of biological research.
• 2. We can learn much about a particular molecule
  by searching databases and using available
  analysis tools
• 3. A large number of databases are available for
  that task. Some databases are very general, some
  are more specialized, while some are very
  specialized. For best results we often need to
  access multiple databases.

31
Biological databases Concluding remarks

• 4. Major types of databases covered in this
  course are focusing on general nucleotide,
  general protein, structure, pathways, molecular
  interactions, protein motifs, publication, and
  specialized databases.
• 5. Common database search methods include keyword
  matching, sequence similarity, motif searching,
  and class searching.
• 6. The problems with using biological databases
  include incomplete information, data spread over
  multiple databases, redundant information,
  various errors, sometimes incorrect links, and
  constant change.

32
Biological databases Concluding remarks

• 7. Database standards, nomenclature, and naming
  conventions are not clearly defined for many
  aspects of biological information. This makes
  information extraction more difficult.
• 8. Retrieval systems help extract rich
  information from multiple databases. Examples
  include Entrez and SRS.
• 9. Formulating queries is a serious issue in
  biological databases. Often the quality of
  results depends on the quality of the queries.

33
Biological databases Concluding remarks

• 10. Statistical measures indicate the quality of
  matches. Often the statistical and biological
  significance are related. Sometimes, however
  matches of real biological significance have low
  statistical scores.
• 11. Access to biological databases is so
  important that today virtually every molecular
  biological project starts and ends with querying
  biological databases.

34
Biological databases Summary of Todays lecture

• Popular databases KEGG, BIND, Pfam, PROSITE,
  PUBMED
• Data retrieval systems Entrez, SRS
• Database searching capability, potential
  problems.
• Statistics
• Protein families (gt 5K)
• Sequence patterns (gt 1.5K)
• Interactions (gt11K or 110 X 110 which is
  relatively few)
• Relatively small amount of data for function
  (e.g. Pathways lt 200)