Biology Reference
In-Depth Information
Methods complementary to SGA include dSLAM
(diploid synthetic lethal analysis by microarray), which
relies on the barcodes associated with each deletion mutant
to enable quantification of double deletion strain abundance
in amixed population
[24]
. Briefly, amarked querymutation
is introduced into a pooled set of heterozygote deletion
strains containing an 'SGAmarker' by mass transformation.
Using the same selection steps as in SGA, double mutant
haploids are selected and the barcode intensities of each
strain in the pool (compared to a non-selected control pool)
provide a measure of relative double mutant fitness. dSLAM
has been applied extensively to map genetic interactions
between genes involved in DNA integrity and histone
modification
[25,26]
. Finally, a third method for genetic
interaction discovery, called genetic interaction mapping
(GIM), was used to examine interactions between genes
involved inmRNA processing
[27]
. GIM represents a hybrid
of SGA and dSLAM. Reminiscent of SGA, double mutants
are generated bymating and sporulation, but as with dSLAM
all steps are performed in a pooled format, which involves
competitive growth of double mutant meiotic progeny, with
interactions identified by comparison of barcode microarray
hybridization intensities between double mutants and
a reference population
[27]
.
or double mutant yeast strains to map a quantitative genetic
interaction network for genes encoding components of the
26S proteasome
[21]
. A similar fluorescence-based assay was
also applied to quantify genetic interactions between dupli-
cated genes
[28]
. Despite providing high-resolution interac-
tion measurements and illustrating the utility of quantitative
genetic interaction analysis for functional analysis of path-
ways and protein complexes, these methods are not easily
amenable to genome-scale studies
[29]
.
The requirement for quantitative phenotypic measure-
ments has imposed constraints on the scale and functional
scope of quantitative genetic interaction studies. However,
large-scale measurement of yeast colony size offers the
potential to identify quantitative genetic interactions on
a scale compatible with the throughput and capacity affor-
ded by methods such as SGA
[17,30]
. In fact, correcting
high-density yeast colony arrays for sources of systematic
variability that plague most, if not all, array-based technol-
ogies resulted in quantitatively accurate single and double
mutant colony size measurements (
Box 6.2
). These
measurements overlap significantly with fitness measure-
ments obtained using other high-resolution methods, indi-
cating that colony size is a suitable proxy for yeast fitness
[17]
. The relative ease with which colony size-based fitness
measurements can be obtained provides a reasonable
compromise between experimental throughput and quanti-
tative resolution. Indeed, a study combining SGA with
a genome-scale colony size scoring methodology examined
~5.4 million gene pairs covering ~30% of the S. cerevisiae
genome
[2]
. This large-scale endeavor measured single and
double mutant yeast fitness to uncover ~170 000 genetic
interactions (~113 000 negative and ~57 000 positive) and
provide the first viewof a quantitative, genome-scale genetic
interaction network for a eukaryotic cell
[2]
.
Quantifying Genetic Interactions
Early genetic interaction studies were predominantly based
on binary (i.e., viable or unviable) assessment of cellular
fitness
[3,4]
. Quantitative measurements enable identification
of subtle negative and positive interactions and the construc-
tion of higher-resolution genetic networks. Most efforts to
map quantitative genetic networks have thus far been based on
the quantitative measurement of cell growth or fitness asso-
ciated with yeast single and double deletion mutants. For
example, a liquid growth profiling approach was used to
quantitatively measure genetic interactions between a subset
of genes involved in DNA replication and repair
[19]
.In
another example, fitness was measured from fluorescence-
labeled populations ofwild-type cellsmixedwith either single
Quantitative Genetic Interaction Profiles
Reveal the Functional Organization of a Cell
The set of synthetic lethal genetic interactions for a given
gene, termed a genetic interaction profile, provides a rich
BOX 6.2 Computational Pipeline for Processing SGA Data
Single and double mutant array plates derived from SGA are
photographed using a high-resolution digital camera. Mutant
array plate images are then processed using custom-developed
image processing software to identify colonies and measure
their area in terms of pixels. To identify quantitative genetic
interactions, yeast colony size pixel data are subjected to
a series of normalization steps to correct for several systematic
experimental effects, after which genetic interactions are
measured by comparing corrected double mutant colony size
to the colony size of the corresponding single mutants. This
analysis generates a genetic profile for each array mutant that
can be used to construct a correlation-based genetic interaction
network (see
Figure 6.1B
and
Figure 6.2
). As an alternative to
fitness-based genetic interactions, mutant array plates may
carry a specific fluorescent reporter that can be analyzed using
a high-content imaging system. Genetic interactions can then
be identified and measured based on various cell biological
parameters and phenotypes. Refer to Baryshnikova et al.
[119]
for detailed protocols and procedures pertaining to genetic
interaction data acquisition, processing and analysis.
