Transcript Increase in complexity in Evolution

Increase in complexity in
evolution (questions, answers,
research programme)
Eörs Szathmáry
Collegium Budapest
Eötvös University Budapest
What are we interested in?
•
•
•
•
•
Genetic basis of organismic complexity
What is organismic complexity?
Complex morphology?
Complex behaviour?
How do you quantify complexity for the
different cases?
• An intuitive feel for complexity is
widespread
Programme complexity
• S (spatial): storage space needed
• T (temporal): execution time
• P (programme): the size of the shortest
programme with given input and output,
given an agreed language
• Partly independent
• Short programmes with complicated
dynamics (chaos, cellular automata)
Complexity II
• Kolmogorov: entirely random sequence has
the highest complexity
• Another problem: in general one cannot
prove that a given programme is the shortest
possible
• A string is random if the minimal
programme producing it is about as long as
the string
• Randomness cannot be contracted
The number of cell types in an
organism (Bonner)
• Countable at our present state of knowledge
• Can be refined with molecular techniques
(microarrays)
• Fits the intuition rather well
• In the animal world there is a correlation
between number of cell types and organism
size, hence between size and complexity
Cell count in a nematode
Bell and Mooers, 1997
Organism size and number of cell
types (Bell)
Does complexity correlate with
the number of genes?
• A few years ago this seemed to be the case
• There is no a priori reason why this should
be so
• Algorithmic complexity: the length of the
minimal programme, written in a specified
language, that solves a particular problem
• Why should tinkered programmes be
minimal?
Genome size and gene number
Genome size and gene number II
Gene number is not so good
• There is a correlation with complexity, but
rather weak…
• Although there is an interesting pattern in
the fraction of genes devoted to various
functions:
Protein functions
Genes for various functions
Interaction density among genes
is better (Szathmáry et al. 2001
Science)
• Cell types need genes to be switched on and
off in an orderly manner
• Genes regulate other genes
• Once a gene is set, this state can be passed
on to offspring in cell division
• Epigenetic inheritance (Jablonka & Lamb,
1995) systems
Complexity related to network
properties of interacting genes?
• Networks are fashionable, but this by itself
does not render them uninteresting
• Other areas in biology have a vast
experience with network properties
• Food web theory in ecology
• Connectance = (number of existing
links)/(number of possible links)
Number of transcriptional
activator families
Egy gén számos más gént
szabályozhat
• Az X gén terméke egy
transzkripciós faktor
• Ez a fehérje az érintett
gének szabályozó
régiójához kötődik
• Aktiválás és gátlás
egyaránt lehetséges
Temporal complexity - yeast
Complexity must be
characterized slightly (?) better
• Delegated complexity: a generative system (genes,
chemistry, language) can be launched with a finite
number of discrete entities
• Immune and nervous systems: excellent examples
• Information carrying capacity of those systems
should be quantified and combined
• Plants do not have a nervous/immune system, they
use secondary metabolites, which must be coded
explicitly (25,498 genes in Arabidopsis)
Increase in genetic complexity
(a) duplication and divergence
(b) symbiosis
(c) epigenesis
Animal phylogeny
* sequenced
genomes
Hox gene duplications
ParaHox evolution
Some vertebrate proteins
assembled from modules