Large-scale RNA-interference (RNAi) screening has become a widely used approach in invertebrate model organisms and in cell culture. RNAi screening has the power to resolve the architecture and dynamic regulation of cellular signalling pathways and can help to identify genetic interactions involved in human diseases 
. RNAi libraries target almost all annotated genes in the human genome and when used in combination with innovative screening technologies allow the analysis of increasingly complex cellular phenotypes. A common assay type, for example in synthetic lethality screening in cancer addresses the viability or fitness of cells. Synthetic lethality occurs when the combination of a mutation in two different genes results in lethality, whereas when either of the genes is mutated, the organism remains viable. The presence of one of these mutations in e.g. in pathophysiologically altered isogenic or recombinant cells but not in normal cells enables identification of genetic interactions with agents – such as RNAi reagents - that mimic the effect of a second genetic mutation 
. Synthetic lethality is indicated by various physiological indicators which are partially and indirectly assessable using fluorescence-, luminescence- or absorbance-based assaying methods. Cellular fitness is often measured by quantifying ATP levels (e.g., CellTiter-Glo), esterase activity and membrane integrity (e.g., Calcein-AM) or by simple cell or nucleus count (e.g. Hoechst DNA stain).
Intracellular ATP [ATP]i serves as an energy carrier that drives virtually all cell functions. Persistent ATP depletion causes a cell to die and in turn cell death is indicated by low ATP levels. Due to its simple accessibility e.g. by an ATP-dependent luciferase-luciferin reaction [ATP]i has been a long-serving indicator of cellular viability. Cellular metabolism creates a continuous demand of energy that requires permanent energy supply. Variation in metabolic activity results in fluctuation of [ATP]i. For example, [ATP]i varies markedly during cell differentiation 
and with circadian rhythm 
. It has been reported that genetically identical eukaryotic cells show significant cell-to-cell variability of cellular mitochondrial mass caused by inhomogeneous distribution of mitochondria during cell division 
, which presumably results in varying [ATP]i between cells of the same population. In order for the cell to keep up with fluctuating energy, it employs different metabolic pathways (protein and DNA synthesis, polysaccharide synthesis, and lipid synthesis) which use different trinucleotides (GTP, UTP, and CTP, respectively) as an energy source 
. While cell death in the long run inevitably results in reduction of [ATP]i, varying [ATP]i is not necessarily an indicator of cell death. Thus, in general [ATP]i is a robust estimator of cell viability, but careful consideration of the mentioned limitations is required when quantification data obtained from [ATP]i measurement are to be analyzed and interpreted.
In healthy cells the cytoplasmic membrane effectively separates the intracellular fluid from the outside environment. It represents an impermeable barrier for charged fluorescent dyes but is permeable for uncharged and hydrophobic compounds such as Calcein acetoxymethyl (AM). Upon permeation of the cytoplasmic membrane, non-fluorescent Calcein-AM is hydrolyzed by intracellular esterases and the product Calcein, a hydrophilic, strongly fluorescent compound remains inside the cell. Thus, Calcein stained cells have esterase activity and in consequence an intact membrane. Due to its low cytotoxicity and simple handling Calcein-AM is a well-suited and commonly employed fluorescent probe for staining viable cells 
. However, Calcein fluorescence intensity only indirectly reflects membrane integrity by cleavage activity of esterases which depends on cellular metabolism and intracellular ATP. These measurements can often further be complicated by extrusion pumps, which remove fluorophores from the cytoplasm 
. Hence, in analogy to measuring intracellular ATP for assessment of cellular viability, analysis and interpretation of quantification data requires careful consideration of the mentioned limitations when cellular viability is assessed using indicators of membrane integrity.
Nucleus or DNA stain using fluorescent molecules, such as Hoechst 33342, Hoechst 33258, DAPI or other dyes have been long-serving and commonly applied indicators of cellular viability. Hoechst 33342 exhibits distinct fluorescence emission upon binding into the minor groove of DNA and stains condensed as well as normal chromatin of living or dead cells. The fluorescence intensity of DNA stain is directly proportional to nuclear DNA content and can therefore be used to monitor DNA replication during cell cycle and cell division. When applied to assess cellular viability the total Hoechst fluorescence signal or the number of cell exhibiting DNA stain is quantified 
. While the simple handling of Hoechst 33342 and similar fluorophores as well as the above described properties of DNA stain enable a broad spectrum of applications, their use in assessment of cellular viability is limited. Due to stoichiometric binding to DNA quantification of total fluorescence signal from cell populations for assessing cellular viability can be misleading. When employed to measure cell number as viability criterion, co-labeling of living and dead cells can yield inaccurate results.
Any of the mentioned approaches yield information on specific cellular conditions and physiological characteristics which only partially and indirectly reflect cellular viability. In fact, none of the described methods clearly and exclusively reflects cellular viability comprehensively. The same is true for other methods assessing cellular viability such as measurement of redox potential by MTT or MTS, the utilization of charged dyes such as trypan blue or propidium iodide or comparable methods. This might explain why striking candidates selected from large-scale screening approaches require re-testing in time-consuming, independently and sequentially conducted secondary experiments to eliminate false-positive hits. To overcome these limitations, we aimed to assess cellular viability by a ‘multiplexing’ approach which combines different methods in a single experiment.
Conventional high-throughput cell-based viability assays, typically run individually and in serial mode, are time and cost intensive and lead to variations. Multiplexing, i.e. the combination of different assaying methods in the same experiment, adds a level of efficiency by reducing sample supply, reducing cell culture and assay consumable requirements. Redundant steps for sample preparation, plate replication, and assay execution are eliminated and experimental variation is reduced.
In this article, we describe a methodology based on a multiplexing approach for synthetic lethality screening to address the issues encountered in the classical assaying methods described above. We aimed to create a viability multiplexing assay which combines multiple indicators in the same experiment, thus decreasing variability and increasing efficiency, throughput and confidence for hit selection. We further aimed to conduct a case study and to measure the quality of the assay using the software web cellHTS2 
. In order to assess the comparability of the employed assaying methods and to calculate false-positive discovery rate, we intended to quantify the overlap of the obtained viability phenotypes.