Detecting and Interpreting Genetic Homology

1
Detecting and InterpretingGenetic Homology
2
Overview

• Definitions and examples of genetic homology
• Inferring homology from sequence similarity using
  FASTA
• Practical homology detection

3
Homology Defined

• "The same organ in different animals under a
  variety of form and function." Sir Richard Owen,
  Lectures on the Comparative Anatomy and
  Physiology of the Invertebrate Animals, 1843.
• "The mechanism of homology is heredity." Allan
  Boyden, Homology and Analogy A century after the
  definitions of "homologue" and "analogue" of
  Richard Owen,1943.
• "Homology is a relation bearing on recency of
  common ancestry." Olivier Rieppel, Homology and
  logical fallacy, 1992.

4
A working definition

• Structures or organs which share a common
  ancestor.

5
Anatomical Homology
6
Sharing a Common AncestorOrthology
7
Protein Orthology
8
Sharing a Common AncestorParalogy
9
Functional Conservation

• Biochemical Function What the protein does on a
  biochemical level, almost always conserved
  between homologues.
• Example Serine/ Threonine Kinase
• Physiological Function What role the protein
  plays within the organism, only conserved for
  orthologous proteins. Example Glycogen Synthase
  Kinase

10
Biochemical Function

• F 1 ATP Synthase
• Main energy conversion motor for all organisms.

11
Sequence Evolution

• Analysis of sequences have revealed certain
  mutational regularities.
• Closely-related Transitions/transversions
• Distantly-related PAM mutation probabilities
• We can use these mutational regularities to help
  us identify distantly-related sequences that we
  would not recognize by visual examination alone.

12
Recognizing ProteinHomology

• Relies primarily on understanding random sequence
  similarity
• Only by knowing what random similarity looks like
  can we tell when two proteins are unusually (or
  significantly) similar,and thus homologous.
• NOTE "Significant Similarity" is not adefinition
  of homology.

13
(No Transcript)
14
Homology Detection Strategy

• Compare sequence of interest ("query") to a
  database of known sequences.
• Tabulate all similarity scores.
• Fit scores to a known distribution to detect
  scores that are significantly greater than the
  mean.

15
Practical Considerations

• What database should I search?
• What kind of sequences should I search with?
• What E-value is significant?
• What can I reliably infer about the function of
  my sequence based on homology?

16
Databases

• Bigger databases have more sequences.
• Bigger databases are also more redundant, which
  can skew the statistics.
• Bigger databases are also poorly annotated
  (homology with an "unidentified sequence doesn't
  really tell you much)
• Bigger databases take lots of time to search.

17
Databases, Part 2

• Smaller databases (like Swiss-Prot) are often
  better curated and annotated.
• Smaller databases are much less redundant.
• Smaller databases can contain phylogenetically
  relevant sequences (all plant or all fungi)
• Smaller databases are much faster to search.

18
Databases Available

• GenBank gt10,000,000 nucleotide sequences (11
  billion nucleotides), very redundant, poorly
  annotated.
• SwissProt 90,000 protein sequences, low
  redundancy, well annotated.
• ESTs Sizes vary, redundancy varies, main
  advantage is large sampling of putatively
  expressed sequences.

19
Whats a significant E-value?

• For a single search, a FASTA E-value of 0.003 is
  significant, though typically quite distant.
• For multiple searches, the E-value cutoff varies
  according to the number of searches.

20
Multiple Database Searches

• 15,000 EST query sequences
• A 0.001 E-value cutoff means that you should
  expect one false positive in 1000 searches.
• Thus with 15,000 searches, we should expect 15
  false positives with a cutoff of 0.001 .
• To reduce the chances of identifying a false
  positive, set the E-value cutoff lower.
• For 15,000 searches, an E-value cutoff of 0.00001
  will mean that you should expect 0.15 false
  positives.

21
Inferring PhysiologicalFunction from Homology
22
Make Predictions

• NXT1 will definitely bind to the nuclear pore
  complex, so is probably involved in nuclear
  transport.
• NXT1 will not bind Ran-GDP, as NTF2 will.
• Conformational changes in Ran upon GTP-binding
  may induce structural changes that allow it to
  bind to NXT1.

23
Summary

• Homologues share a common ancestor.
• Genetic homology is inferred from significant
  similarity.
• Biochemical function can be reliably inferred
  from genetic homology physiological function
  cannot.