Ohnos Evolution by Duplication and DNA Correlations

1
Ohnos Evolution by Duplication andDNA
Correlations

• Wentian Li, Ph.D
• The Robert S Boas Center for Genomics and Human
  Genetics
• North Shore LIJ Institute for Medical Research

March 18, 2005
2
Outline of the talk

• Ohnos evolution by gene duplication
• Duplication-mutation model (expansion-modification
  model)
• Confirmation of 1/f noise in DNA sequences as
  predicted by the duplication-mutation model
• Duplication in bacteria genomes?
• Music as the second example where the three
  themes converge (1/f noise - redundancy in
  musical composition - duplication-mutation model)

3
Susumu Ohno (1928-2000)

• Undergraduate degree in veterinary (1949)
• Graduate degree in immunology (1953)
• UCLA, then City of Hope National Medical Center
  (50s-retired)
• One X-chromosome being heterochromatic(1959)
• Evolution by gene duplication (1970)
• Term junk DNA (1972)
• Relating DNA with music (1986)

4
From Evolution by Gene Duplication (1970)

• Had evolution been entirely dependent upon
  natural selection, from a bacterium only numerous
  forms of bacteria would have emerged. The
  creation of metazoans, vertebrates, and finally
  mammals from unicellular organisms would have
  been quite impossible, for such big leaps in
  evolution required the creation of new gene loci
  with previously nonexistent function. Only the
  cistron gene that became redundant was able to
  escape from the relentless pressure of natural
  selection. By escaping, it accumulated formerly
  forbidden mutations to emerge as a new gene
  locus.

5
Natural selection merely modified while
redundancy created

• -S. Ohno

6
Pre-1970 works on gene duplication

• Kuwada (1911), Tischler (1915)
• JBS Haldane (1932), The Causes of Evolution
  (Harper and Bros)
• Bridgess observation of gene duplication in
  Drosophila (1935-36)
• Serebrovsky (1938) selection is relaxed in genes
  that duplicate
• Muntzing (1936), Tischler (1935), Nishiyama
  (1934)
• Gulick (1944) increases in gene count.. (to)
  great complexity
• Goldschmit (1940), Metz(1947), Huxley (1942)
  importance of..
• SG Stephens (1951), Possible significance of
  duplication in evolution, Adv. Genetics,
  4247-265.
• Lewis (1951) pseudoallelism gene evolution
  linked duplicates
• S Ohno, U Wolf, NB Atkin (1968), Evolution from
  fish to mammals by gene duplication, Hereditas,
  59169-187.
• See Taylor Raes (2004)

7
More and more interests in gene duplications

• http//www.nslij-genetics.org/duplicaiton/

8
Example two rounds of duplication (2R)

9
The extent of duplicated genes (Dujon et al.
Nature 2004)
10
effects of duplicated genes vs. single-copy genes

• Gu, et al. Nature, 42163-66 (2003)
• Yeast S. cerevisiae has 6000 genes, about ¼ of
  them are part of duplicates, the rest exist as
  singletons.
• Knocking 12.4 of one of the duplicates is
  lethal
• Knocking 29 of the singletons is lethal
• Knocking 64.3 of one of the duplicates has no
  or weak effect
• Knocking 39.5 of the duplicates has no or weak
  effect

11
Could the result be relevant to human genome?

• Housekeeping genes more likely in duplicate
  group (multigene families)?
• Disease genes for Mendelian diseases more likely
  in singleton group??
• Disease genes for complex diseases more likely in
  duplicate group???

12
Some examples of disease genes
13
the general principle of Ohnos duplication is
true, but

• Whole genome duplication (polyploidization) or
  regional/local/segmental duplication?
• polyploids tend to be unstable
• Duplication of genes or duplication of a piece of
  DNA with possibly no functions?
• then gene duplication is a by-product
  instead of a design
• Is redundancy caused by duplication the only
  source of robustness against mutations?
• system robustness is another source

14
Redundancy and correlation

• Redundancy, n. repetition or excessive use of
  something
• 1. Correlation, n. a causal, complementary,
  parallel or reciprocal relationship between two
  comparable entities
• 2. Correlation, n. (statistics) the
  simultaneous change of two random variables

• The Copy and the original gene (DNA segment) are
  correlated in sense 1.
• Even after the decay and deletion of function of
  the copy gene, it is still correlated with the
  original in sense 2. problem not random
  variable

15
Modeling correlation via duplication

• Duplication-mutation model (expansion-modification
  model)
• W.Li, Europhysics Letters,10395-400 (1989)
  
• W.Li, Physical Review A, 435240-6260 (1991)
• Repeated local duplication with a fixed
  probability (p)
• Repeated mutation with a fixed probability
  (q1-p)
• Sequence length continues to increase

16
Properties of the duplication-mutation model

• Although the duplication/redundancy/correlation
  is created locally, the correlation can actually
  be propagated to long distances
• Duplication part is responsible for generating
  correlation, whereas the mutation part is
  destroying it.
• When the duplication probability is large (e.g.
  p0.9, q0.1), the limiting sequence exhibits
  long-range correlation and power-law (scaling)
  behaviors, including the 1/f noise.

17
History behind the duplication-mutation model

• Searching for long-range correlation in limit
  sequences generated by cellular automata (1986)
• The idea of increasing the sequence length in
  order to generate long-range correlation (1988)
• Realizing the connection between the
  duplication-mutation model (expansion-modification
  model) and Ohnos evolution by duplication (K
  Kaneko, Nov 1988)
• Failed attempt to detect 1/f noise in DNA
  sequences from GeneBank (Li, Kaneko, 1989)
• Successful attempt to detect 1/f noise in DNA
  sequences (Li, 1991)

18
duplication
19
(No Transcript)
20
(No Transcript)
21
(No Transcript)
22
(No Transcript)
23
(No Transcript)
24
Li, Stolovitzky, Bernaola-Galvan, Oliver, Genome
Research, 8916-928 (1998)
25
(No Transcript)
26
Li and Holste, Fluctuation and Noise Letters,
4L453-L464 (2004)
27
(No Transcript)
28
Li and Holste, Phys. Rev. E (2005), to appear.
29
(No Transcript)
30
The general idea in duplication-mutation model
seems to fit the data, but

• The local duplication and the subsequent
  pushing may be replaced by a global duplication
  (I.e. base duplication may be replaced by
  segmental duplication).
• Whole-genome duplication followed by several
  inter-chromosome exchanges may also create
  redundancy and correlation
• A spectrum of dynamics instead of just two
  (duplication and mutation)

31
Duplication in bacteria genomes?

• Zipkas Riley (1975) pair of genes of similar
  function tend to be separated by 90 or 180
  degrees
• Lobry (1996) G-C changes sign at replication
  origin/termina
• Eisen et al (2000) when two bacterial genomes
  are compared, either direct or complementary
  opposite, there is a cross-like match

32
If GC-bias changes sign but has the same
magnitude, two similar sequences point to
opposite direction
33
(No Transcript)
34
(No Transcript)
35
(No Transcript)
36
Genomewide dot plots

• Base-to-base comparison
• Base composition difference
• Oligonucleotide composition differences
• Exact match of a long (e.g. 25 bases) oligo
• Protein/ORF matches (BLAST, FASTA)
• maximum unique match (MUM) (MUMer program
  http//www.tigr.org/software/mummer/)

37
(No Transcript)
38
The all pervasive principle of repetitious
recurrence governs not only coding sequence
construction but also human endeavor in musical
composition

• Title of a paper authored by Susumu Ohno and
  Midori Ohno (1996)

39
DNA-Music connection I

• DNA sequences all exhibit 1/f spectra (no
  exception yet)
• Musical time series all exhibit 1/f noise (both
  loudness and pitch, both musical signal and
  speech)
• Voss Clarke (1975) 1/f noise in music and
  speech, Nature, 258317-318.

40
DNA-Music connection II

• DNA sequence is redundant (full of repeats)
• Musical series is also redundant (repetitive)

41
DNA-Music connection III

• The generation/elongation of DNA sequences are
  driven mainly by duplication in various form
  (genomewide, segmental,)
• The musical composition process consists of
  re-usage of the main/minor themes
• Can both be modeled by some form of
  duplication-mutation models?

42
We have formerly seen that parts many times
repeated are eminently liable to vary in number
and structure consequently it is quite probable
that natural selection, during the long-continued
course of modification, should have seized on a
certain number of the primordially similar
elements, many times repeated, and have adapted
them to the most diverse purposes.

• Charles Darwin, 1859 (The Origin of Species, page
  477)