robustness - senecio7の日記

Papp B, Notebaart RA & Pál C 2011 Systems-biology approaches for predicting genomic evolution. Nat Rev Genet 12:591-602.

is evolution predictable at the molecular level?
this Review investigates recent progresses in mapping evolutionary trajectories and discusses the degree to which these predictions are realistic

evolutionary biology successfully interprets molecular and cellular phenotypes as a result of diverse evolutionary forces that acted in the past
it rarely builds an explicit theoretical framework to predict potential routes of evolution
it is unclear how predictable evolution is at the level of genomes and molecular networks
such a framework has the potential to permit informed decisions in medicine⁴, biotechnology⁵ and environmental issues⁶
we demonstrate that it is possible to predict, rather than simply interpret, past evolution by synthesizing evolutionary theory, systems biology and molecular data

there are three layers of prediction that we consider in this Review
predicting the distribution of mutational effects and epistasis
that is, parameters that influence many central issues in evolutionary genomics
explaining the driving forces of sequence and expression evolution on a genomic scale
understanding why particular evolutionary trajectories are realized whereas others are not

models that are most suited for evolutionary studies are based on sound biochemical principles
they should capture the functional states of the cell and compute phenotypes (for example, growth rate) that serve as fitness correlates

mutations with weak phenotypic effects are common
epistasis between mutation is wide-spread
it remains largely unexplored in what extent the distribution of epistatic interactions changes across loci and environmental conditions

nearly 80% of protein-coding genes in Saccharomyces cerevisiae are not essential for viability under standard laboratory conditions
these findings raise questions about the mechanistic basis of gene dispensability and about whether this tolerance to inactivation is the result of an evolved capacity of genetic networks to compensate for mutations
these genes are important under other natural environmental settings that are not yet investigated in the laboratory
gene deletions may be compensated for by a gene duplicate with a redundant function
reorganization of metabolic fluxes across alternative pathways⁴² may buffer gene loss

a computational analysis of S. cerevisiae metabolism showed that a large fraction of non-essential enzyme-encoding genes catalyse reactions that are inactive under the tested condition
by simulating gene deletions under several different nutrient condition, the model claimed that many functionally inactive genes would become essential under some other conditions
most of the apparently dispensable genes (37–68%) belong to this category
redundant gene duplicates (15–28%) and alternative pathways (4–17%) can only explain a few cases
these condition-specific genes have limited phylogenetic distribution

approximately 50% of reactions are estimated to be inactive under laboratory conditions
redundancy through duplicate genes was the major (37.5%) molecular mechanism behind gene dispensability
alternative pathways constituted the minor mechanisms (12.5%)

97% of the single-gene deletions exhibit a measurable growth phenotype in at least one of the hundreds of tested conditions

environmental specificity is the dominant explanation for apparent dispensability

most synthetic lethal interactions between loci are restricted to certain environments owing to lack of compensation under some nutrient conditions
robustness against null mutations is unlikely to be a directly selected trait
it is a side effect of adaptation to survive in changing conditions

the distribution of epistatic interactions across pairs of loci is clearly non-uniform and probably trimodal
pairs of loci show either very strong negative or positive interactions or no interactions at all
epistatic interactions are generally parasitic across environmental conditions
evolutionary theories that rely on a constant adaptive landscape (for example, theories on the origins of robustness against mutations or the impact of compensatory mutations on evolution) are far too simplistic
a few genes ('hubs') exhibit an especially large number of epistatic interactions
the majority of genes display few interactions
there is a strong link between the degree of epistatic interaction and the extent of pleiotropy
hub genes are likely to influence the phenotypic effects of mutations in many other genes
their loss may markedly alter which mutations paths are available for adaptation
epistatic interactions are frequent between biochemical pathways or modules
a module should be tightly integrated by strong pleiotropic effects and should be largely independent from other such modules
the claim that modularity facilitates evolution by minimizing pleiotropic constraints⁴⁸ needs to be re-evaluated

the emerging field of evolutionary systems biology reinvestigates central issues in evolutionary biology by using realistic and organism-specific models of cellular subsystems
they calculate evolutionarily relevant variables that are difficult to estimate experimentally on a large-scale or across environmental conditions
these models also hold the promise to transform evolutionary biology into a more predictive discipline
systems biology and population genetics should calculate the mutational effects and fixation rate of mutations, respectively

neutral evolution and emergence of key innovations
how does the neutral evolution of metabolic networks influence the emergence of evolutionary innovations⁹¹?
the presence of alternative metabolic circuits with the same phenotype is a key facilitator of evolutionary novelty

importance of regulatory versus structural mutations in adaptive evolution
phenotypic changes could arise through mutations in cis-regulatory sequences or coding regions
their relative importance remains intensely debated

reverse ecology aims to gain insights into the habitats in which organisms have evolved based on comparison of networks across a wide range of species
comparative systems-biology models could provide enormous help in the identification of new drug targets shared by numerous related pathogenic bacteria