Biology Reference
In-Depth Information
DEFINING GENETIC INTERACTIONS
Geneticists have long recognized that genetic interactions
are important for shaping the phenotypic landscape of
a population. In 1909 William Bateson introduced the term
'epistasis' (i.e., 'standing upon') to describe a specific type
of genetic interaction whereby one mutation masks the
effects of another mutation on fur color in rabbits and mice
[8] ( Figure 6.1A ). The same term was later adopted to
describe genetic relationships that often occur among
members of the same metabolic pathways, where the action
of one enzyme depends on a substrate produced by another
enzyme ( Figure 6.1A ). Ronald Fisher later expanded this
term to include any multi-locus mutant effect that deviates
from the additive combination of the corresponding indi-
vidual loci [9] . Today, Fisher's epistasis is often used to
generally define a genetic interaction as an unexpected
phenotype that cannot be explained by the combined effects
of the individual mutations ( Figure 6.1A ).
Adopting this definition, a genetic interaction between
two genes can be experimentally defined based on measure-
ment and comparison of three fundamental properties that
include the single mutant phenotypes, an estimate of the
expected doublemutant phenotype, and ameasurement of the
observed double mutant phenotype ( Figure 6.1A ). Although
these are relatively straightforward criteria, determining how
mutations in different genes are expected to combine is not as
obvious. The additive and multiplicative models of genetic
interactions represent the two most common approaches for
measuring non-independence of gene effects on phenotypic
variance [10,11] . In the case of the additive model, the
phenotypes associatedwithmutant alleles of individual genes
are expected to combine, such that the double mutant
phenotype should be equal to the sumof the two singlemutant
phenotypes. Alternatively, according to the multiplicative
model, genes are expected to change a phenotype by a specific
fraction, and thus a double mutant phenotype should be equal
to the product of the two single mutants phenotypes. The
choice of model is dependent on the particular phenotype and
scale of measurement (e.g., linear vs. logarithmic) because
genes may appear independent on one scale but show
a genetic interaction if measured on a different scale [10,11] .
In the case of yeast fitness, the expected double mutant
phenotype is typically modeled as a multiplicative combi-
nation of single mutant phenotypes and genetic interactions
are measured by the extent to which double mutants deviate
from this multiplicative expectation [10,11] ( Figure 6.1A ).
Negative Genetic Interactions
Negative genetic interactions describe double mutants that
exhibit a more severe phenotype than expected [10,11]
( Figure 6.1A ). Synthetic lethality represents an extreme
negative interactionwhere twomutations, each causing little
fitness defect on their own, result in an unviable phenotype
when combined as double mutants. This phenomenon,
initially observed among the progeny of intercrosses
between natural variants of Drosophila pseudoobscura,
provided the first insight into the degree of genetic vari-
ability concealed within natural populations [12] . Synthetic
lethality has since been explored extensively in yeast and
proven to be an extremely powerful genetic tool for identi-
fying and characterizing genes in almost all biological
processes [13
15] . Negative genetic interactions, such as
synthetic sickness or synthetic lethality, are interesting
because they often occur between genes that impinge on
a common essential biological function [4] . Alternatively,
genes functioning in the same essential pathway or complex
may also share a negative genetic interaction if each muta-
tion has adverse effects on pathway activity [16
e
18] .
e
Positive Genetic Interactions
Positive interactions describe double mutants exhibiting
a less severe phenotype than expected based on the product
of the two single mutant phenotypes, and can be further
sub-classified into a variety of categories ( Figure 6.1A ) [11,
19,20] . For example, the 'symmetric' category describes
a type of positive interaction whereby the phenotypes
associated with the single and the resultant double mutant
are quantitatively indistinguishable [19,20] . Conversely,
the 'asymmetric' class consists of those interactions in
which the strength of the phenotypic effect varies between
single and double mutants [19,20] .
Importantly, these types of positive genetic interaction
are often associated with different biological interpretations
and, when measured quantitatively, offer the potential to
infer biochemical relationships between gene products and
elucidate how biological pathways and complexes relate to
one another to modulate cellular functions [19
e
21] .For
mutant fitness is greater than the expected double mutant fitness but less than or equal to the fitness of the sickest single mutant. Suppression occurs when
the observed double mutant fitness is greater than the fitness of the sickest single mutant. In the case of deletion mutants, suppression is often indicative of
a negative regulatory relationship between the interacting genes. (B) The set of negative and interactions for a given mutant is referred to as a genetic
interaction profile. Genes that share genetic interactions in common tend to be functionally related. Two-dimensional hierarchical clustering is an effective
way to group genes together based on genetic interaction profile similarity. Arranging genes in this manner enables identification of highly correlated
groups of genes that correspond to functionally related gene modules, including protein complexes and pathways, as well as novel functional predictions.
In addition to clustering algorithms, genetic interactions can also be visualized as a correlation-based network connecting genes with similar genetic
interaction profiles. Using a force-directed network layout, genes with highly similar genetic interaction profiles (black lines) are placed close to each
other in the network, while genes with less similar interaction profiles (blue lines) are place further apart from one another.
Search WWH ::




Custom Search