A complete mass-spectrometric map of the yeast proteome applied to quantitative trait analysis

Experience from different fields of life sciences suggests that accessible, complete reference maps of the components of the system under study are highly beneficial research tools. Examples of such maps include libraries of the spectroscopic properties of molecules, or databases of drug structures in analytical or forensic chemistry. Such maps, and methods to navigate them, constitute reliable assays to probe any sample for the presence and amount of molecules contained in the map. So far, attempts to generate such maps for any proteome have failed to reach complete proteome coverage. Here we use a strategy based on high-throughput peptide synthesis and mass spectrometry to generate an almost complete reference map (97% of the genome-predicted proteins) of the Saccharomyces cerevisiae proteome. We generated two versions of this mass-spectrometric map, one supporting discovery-driven (shotgun) and the other supporting hypothesis-driven (targeted) proteomic measurements. Together, the two versions of the map constitute a complete set of proteomic assays to support most studies performed with contemporary proteomic technologies. To show the utility of the maps, we applied them to a protein quantitative trait locus (QTL) analysis, which requires precise measurement of the same set of peptides over a large number of samples. Protein measurements over 78 S. cerevisiae strains revealed a complex relationship between independent genetic loci, influencing the levels of related proteins. Our results suggest that selective pressure favours the acquisition of sets of polymorphisms that adapt protein levels but also maintain the stoichiometry of functionally related pathway members.

Effect of pasteurization on the protein composition and oxidative stability of beer during storage

The impacts of pasteurization of a lager beer on protein composition and the oxidative stability were studied during storage at 22 °C for 426 days in the dark. Pasteurization clearly improved the oxidative stability of beer determined by ESR spectroscopy, whereas it had a minor negative effect on the volatile profile by increasing volatile compounds that is generally associated with heat treatment and a loss of fruity ester aroma. A faster rate of radical formation in unpasteurized beer was consistent with a faster consumption of sulfite. Beer proteins in the unpasteurized beer were more degraded, most likely due to proteolytic enzyme activity of yeast remnants and more precipitation of proteins was also observed. The differences in soluble protein content and composition are suggested to result in differences in the contents of prooxidative metals as a consequence of the proteins ability to bind metals. This also contributes to the differences in oxidative stabilities of the beers.

A surface-activated chemical ionization approach allows quantitative phosphorylation analysis of the cyclin-dependent kinase inhibitor Sic1 phosphorylated on Ser201

RATIONALE

Quantitative phosphoproteomics represents a front line for functional proteomics and hence for systems biology. Here we present a new application of the surface-activated chemical ionization (SACI) technology for quantitative phosphoproteomics analysis. The main advantages of SACI-MS technology are high sensitivity, quantitative accuracy and matrix effect reduction, which allow quantitative estimations.

METHODS

A SACI-MS approach was used to investigate the quantitative in vivo phosphorylation of the cyclin-dependent kinase inhibitor Sic1, a low-abundance protein of Saccharomyces cerevisiae, which is phosphorylated on Ser201 by casein kinase 2 (CK2) and compared its phosphorylation status in cells growing in two different carbon sources (glucose or ethanol).

RESULTS

Our relative quantification indicated that the Sic1-Ser201 phosphorylation level is about 2-fold higher in ethanol- than in glucose-growing cells, proportional to the Sic1 protein level. This finding is coherent with results of western blot analysis using anti-phospho-Ser201-specific antibody, validating the results obtained with this new SACI approach.

CONCLUSIONS

The findings presented in this paper indicate that the innovative LC/SACI-MS method, coupled with immunoprecipitation, is a powerful device to obtain quantitative information on the phosphorylation state of low abundance proteins.

In vivo determination of organellar pH using a universal wavelength-based confocal microscopy approach

Many essential cellular processes are affected by transmembrane H(+) gradients and intracellular pH (pHi). The research of such metabolic events calls for a non-invasive method to monitor pHi within individual subcellular compartments. We present a novel confocal microscopy approach for the determination of organellar pHi in living cells expressing pH-dependent ratiometric fluorescent proteins. Unlike conventional intensity-based fluorometry, our method relies on emission wavelength scans at single-organelle resolution to produce wavelength-based pH estimates both accurate and robust to low-signal artifacts. Analyses of Ato1p-pHluorin and Ato1p-mCherry yeast cells revealed previously unreported wavelength shifts in pHluorin emission which, together with ratiometric mCherry, allowed for high-precision quantification of actual physiological pH values and evidenced dynamic pHi changes throughout the different stages of yeast colony development. Additionally, comparative pH quantification of Ato1p-pHluorin and Met17p-pHluorin cells implied the existence of a significant pHi gradient between peripheral and internal cytoplasm of cells from colonies occurring in the ammonia-producing alkali developmental phase. Results represent a step forward in the study of pHi regulation and subcellular metabolic functions beyond the scope of this study.

Perfluorooctanoic acid and ammonium perfluorooctanoate: volatile surfactants for proteome analysis?

RATIONALE

Fluorinated surfactants are being explored as mass spectrometry (MS)-friendly alternatives to sodium dodecyl sulfate (SDS) for proteome analysis. Previous work demonstrates perfluorooctanoic acid (PFOA) to be compatible with electrospray ionization (ESI)-MS. The high volatility of PFOA provides an intrinsic approach to potentially eliminate the surfactant during ESI, or alternatively through solvent evaporation prior to MS. The ammonium salt of PFOA, ammonium perfluorooctanoate (APFO), is likely favored for proteome experiments; the MS and liquid chromatography (LC)/MS tolerance of APFO has not been established for proteome applications.

METHODS

Standard proteins and peptides, as well as a yeast proteome mixture, were individually spiked with surfactants (APFO, PFOA, SDS), and subjected to direct infusion ESI-MS, LC/MS/MS and LC/UV. The level of fluorinated surfactant remaining after solvent evaporation under varying conditions (time, pH, salt and protein content) was quantified and compared to the threshold tolerance level of the surfactant in an MS experiment (determined herein).

RESULTS

Whereas PFOA is found ineffective at assisting protein solubilization, APFO is as effective as SDS for resolubilization of acetone-precipitated yeast proteins (~100% recovery). Unfortunately, the LC and MS threshold tolerance of APFO is only minimally greater than SDS (~2-fold higher concentration to cause 50% suppression). Nonetheless, the benefits of APFO in a proteome experiment are realized following a one-step evaporation protocol for removal of the surfactant in acidified solvent.

CONCLUSIONS

APFO is considered a favoured alternative to SDS for proteome solubilization. Strictly speaking, APFO is not an 'MS-friendly' surfactant for proteome characterization; the detergent not only suppresses ESI signals at high concentration, but also perturbs reversed phase separation. However, the simplicity of APFO removal ahead of LC/MS justifies its use over the conventional SDS.

Capillary electrophoresis with Orbitrap-Velos mass spectrometry detection

Capillary electrophoresis can provide fast and efficient separations of peptides. However, the high speed separation and limited loading capacity of capillary electrophoresis requires the use of a fast and sensitive detector. While laser-induced fluorescence provides exquisite sensitivity and millisecond response time, it inherently generates a low information content signal. In contrast, mass spectrometry provides an information rich signal that is attractive for peptide analysis. The recently introduced Velos-Orbitrap mass spectrometer is capable of fast and sensitive tandem MS acquisition and simultaneous high accuracy MS acquisition, which is well suited for coupling with fast and efficient separation methods for peptide analysis. We evaluated this instrument as a detector for peptide separation by capillary electrophoresis. In MS mode, we observed low attomole detection limits for a number of peptides in a tryptic digest of standard proteins with high mass resolution (30,000 at m/z 400). The response time of the Orbitrap at this resolution was ∼0.70s, which was adequate to reconstruct the peak shape and area of our electrophoretic peaks. The linear ion-trap successfully recorded tandem MS spectra of tryptic peptides at 20 nM concentration.

The effect of selenium enrichment on baker's yeast proteome

The use of regular yeast (RY) and selenium-enriched yeast (SEY) as dietary supplement is of interest because the Nutritional Prevention of Cancer (NPC) trial revealed that SEY but not RY decreased the incidence of prostate cancer (PC). Using two-dimensional difference in gel electrophoresis (2D-DIGE)-tandem mass spectrometry (MS/MS) approach, we performed proteomic analysis of RY and SEY to identify proteins that are differentially expressed as a result of selenium enrichment. 2D-DIGE revealed 96 candidate protein spots that were differentially expressed (p≤0.05) between SEY and RY. The 96 spots were selected, sequenced by LC/MS/MS and 37 proteins were unequivocally identified. The 37 identified proteins were verified with ProteinProphet software and mapped to existing Gene Ontology categories. Furthermore, the expression profile of 5 of the proteins with validated or putative roles in the carcinogenesis process, and for which antibodies against human forms of the proteins are available commercially was verified by western analysis. This study provides evidence for the first time that SEY contains higher levels of Pyruvate Kinase, HSP70, and Elongation factor 2 and lower levels of Eukaryotic Translation Initiation Factor 5A-2 and Triosephosphate Isomerase than those found in RY.

Multiple TORC1-associated proteins regulate nitrogen starvation-dependent cellular differentiation in Saccharomyces cerevisiae

BACKGROUND

The budding yeast Saccharomyces cerevisiae undergoes differentiation into filamentous-like forms and invades the growth medium as a foraging response to nutrient and environmental stresses. These developmental responses are under the downstream control of effectors regulated by the cAMP/PKA and MAPK pathways. However, the upstream sensors and signals that induce filamentous growth through these signaling pathways are not fully understood. Herein, through a biochemical purification of the yeast TORC1 (Target of Rapamycin Complex 1), we identify several proteins implicated in yeast filamentous growth that directly associate with the TORC1 and investigate their roles in nitrogen starvation-dependent or independent differentiation in yeast.

METHODOLOGY

We isolated the endogenous TORC1 by purifying tagged, endogenous Kog1p, and identified associated proteins by mass spectrometry. We established invasive and pseudohyphal growth conditions in two S. cerevisiae genetic backgrounds (Σ1278b and CEN.PK). Using wild type and mutant strains from these genetic backgrounds, we investigated the roles of TORC1 and associated proteins in nitrogen starvation-dependent diploid pseudohyphal growth as well as nitrogen starvation-independent haploid invasive growth.

CONCLUSIONS

We show that several proteins identified as associated with the TORC1 are important for nitrogen starvation-dependent diploid pseudohyphal growth. In contrast, invasive growth due to other nutritional stresses was generally not affected in mutant strains of these TORC1-associated proteins. Our studies suggest a role for TORC1 in yeast differentiation upon nitrogen starvation. Our studies also suggest the CEN.PK strain background of S. cerevisiae may be particularly useful for investigations of nitrogen starvation-induced diploid pseudohyphal growth.

The FN3 and BRCT motifs in the exomer component Chs5p define a conserved module that is necessary and sufficient for its function

Chs5p is a component of the exomer, a coat complex required to transport the chitin synthase Chs3p from the trans-Golgi network to the plasma membrane. The Chs5p N-terminal region exhibits fibronectin type III (FN3) and BRCT domains. FN3 domains are present in proteins that mediate adhesion processes, whereas BRCT domains are involved in DNA repair. Several fungi--including Schizosaccharomyces pombe, which has no detectable amounts of chitin--have proteins similar to Chs5p. Here we show that the FN3 and BRCT motifs in Chs5p behave as a module that is necessary and sufficient for Chs5p localization and for cargo delivery. The N-terminal regions of S. cerevisiae Chs5p and S. pombe Cfr1p are interchangeable in terms of Golgi localization, but not in terms of exomer assembly, showing that the conserved function of this module is protein retention in this organelle and that the interaction between the exomer components is organism-specific.

Monitoring cytoplasmic protein complexes with blue native gel electrophoresis and stable isotope labelling with amino acids in cell culture: analysis of changes in the 20S proteasome

Analysis of protein complexes is of increasing interest in the field of proteomics. A challenge is to develop methods for monitoring changes in the quantity and subunit composition of protein complexes on a proteome-wide scale. Here, we describe the combination of 1-D blue native polyacrylamide gel electrophoresis (BN-PAGE) with stable isotope labelling of amino acids in cell culture (SILAC) and tandem mass spectrometry (MS/MS). Cleared lysates from normal and perturbed samples, one incorporating heavy stable isotopes and the other light isotopes, are co-separated by blue native PAGE and then analysed and quantitated with MS/MS and appropriate software. This permits the analysis of cytoplasmic complexes. To demonstrate this technique, we explored how the 20S proteasome changes when the Pre9/α3 subunit, the only non-essential subunit of this complex, was deleted. Our results showed that ΔPre9/α3 cells can form the 20S proteasome complex, although with reduced efficiency. This involves an increase in expression of the α4 subunit. Our findings suggest this technique as an approach for the study of quantitative and qualitative differences in protein complexes, from cleared cell lysates.

Influence of yeast macromolecules on sweetness in dry wines: role of the saccharomyces cerevisiae protein Hsp12

Yeast autolysis during lees contact influences the organoleptic properties of wines especially by increasing their sweet taste. Although observed by winemakers, this phenomenon is poorly explained in enology. Moreover, the compounds responsible for sweetness in wine remain unidentified. This work provides new insights in this way by combining sensorial, biochemical and genetic approaches. First, we verified by sensory analysis that yeast autolysis in red wine has a significant effect on sweetness. Moderate additions of ethanol or glycerol did not have the same effect. Second, a sapid fraction was isolated from lees extracts by successive ultrafiltrations and HPLC purifications. Using nano-LC-MS/MS, peptides released by the yeast heat shock protein Hsp12p were distinctly identified in this sample. Third, we confirmed the sweet contribution of this protein by sensorial comparison of red wines incubated with two kinds of yeast strains: a wild-type strain containing the native Hsp12p and a deletion mutant strain that lacks the Hsp12p protein (Δ°HSP12 strain). Red wines incubated with wild-type strain showed a significantly higher sweetness than control wines incubated with Δ°HSP12 strains. These results demonstrated the contribution of protein Hsp12p in the sweet perception consecutive to yeast autolysis in wine.

Single-particle tracking of quantum dot-conjugated prion proteins inside yeast cells

Yeast is a model eukaryote with a variety of biological resources. Here we developed a method to track a quantum dot (QD)-conjugated protein in the budding yeast Saccharomyces cerevisiae. We chemically conjugated QDs with the yeast prion Sup35, incorporated them into yeast spheroplasts, and tracked the motions by conventional two-dimensional or three-dimensional tracking microscopy. The method paves the way toward the individual tracking of proteins of interest inside living yeast cells.

Protein-fragment complementation assays for large-scale analysis, functional dissection and dynamic studies of protein-protein interactions in living cells

Protein-fragment Complementation Assays (PCAs) are a family of assays for detecting protein-protein interactions (PPIs) that have been developed to provide simple and direct ways to study PPIs in any living cell, multicellular organism, or in vitro. PCAs can be used to detect PPI between proteins of any molecular weight and expressed at their endogenous levels. Proteins are expressed in their appropriate cellular compartments and can undergo any posttranslational modification or degradation that, barring effects of the PCA fragment fusion, they would normally undergo. Assays can be performed in any cell type or model organism that can be transformed or transfected with gene expression DNA constructs. Here we focus on recent applications of PCA in the budding yeast, Saccharomyces cerevisiae, that cover the gamut of applications one could envision for studying any aspect of PPIs. We present detailed protocols for large-scale analysis of PPIs with the survival-selection dihydrofolate reductase (DHFR), reporter PCA, and a new PCA based on a yeast cytosine deaminase reporter that allows for both survival and death selection. This PCA should prove a powerful way to dissect PPIs. We then present methods to study spatial localization and dynamics of PPIs based on fluorescent protein reporter PCAs.

Software pipeline and data analysis for MS/MS proteomics: the trans-proteomic pipeline

The LC-MS/MS shotgun proteomics workflow is widely used to identify and quantify sample peptides and proteins. The technique, however, presents a number of challenges for large-scale use, including the diverse raw data file formats output by mass spectrometers, the large false positive rate among peptide assignments to MS/MS spectra, and the loss of connectivity between identified peptides and the sample proteins that gave rise to them. Here we describe the Trans-Proteomic Pipeline, a freely available open source software suite that provides uniform analysis of LC-MS/MS data from raw data to quantified sample proteins. In a straightforward manner, users can extract MS/MS information from raw data of many instrument formats, submit them to search engines for peptide identification, validate the results to remove false hits, combine together results of multiple search engines, infer sample proteins that gave rise to the identified peptides, and perform quantitation at the peptide and protein levels.

The effect of interfering ions on search algorithm performance for electron-transfer dissociation data

Collision-activated dissociation and electron-transfer dissociation (ETD) each produce spectra containing unique features. Though several database search algorithms (e.g. SEQUEST, MASCOT, and Open Mass Spectrometry Search Algorithm) have been modified to search ETD data, this consists chiefly of the ability to search for c- and z(*)-ions; additional ETD-specific features are often unaccounted for and may hinder identification. Removal of these features via spectral processing increased total search sensitivity by approximately 20% for both human and yeast data sets; unique peptide identifications increased by approximately 17% for the yeast data sets and approximately 16% for the human data set.

Split-EGFP screens for the detection and localisation of protein-protein interactions in living yeast cells

Proteomics aims to identify and classify the proteins present in a particular cell or tissue. However, we know that proteins rarely function alone and knowledge of which proteins interact with which other proteins is vital if we wish to understand how cells work. The budding yeast, Saccharomyces cerevisiae, is a well-established model for studying protein-protein interactions, and a number of methods have been developed to do this. A method for the in vivo detection and localisation of interacting pairs of proteins in living yeast cells is presented. The method relies on the ability of fragments of enhanced green fluorescent protein (EGFP) to reassemble if brought into close proximity. The reassembled EGFP regains the ability to fluoresce, and this fluorescence can be detected providing evidence of interaction and information about its location. S. cerevisiae is an ideal organism to apply this method to due to the relative ease with which its genome can be manipulated. The method described enables the modification of S. cerevisiae genes at the 3'-end with DNA encoding fragments of EGFP. Consequently, the expression levels of the proteins are unlikely to be affected and thus the method is unlikely to result in false positives. In addition to the protocol for labelling and detection of interacting pairs of yeast proteins, methods for simple tests for the effects of the labelling on the organism's function are presented.

Full dynamic range proteome analysis of S. cerevisiae by targeted proteomics

The rise of systems biology implied a growing demand for highly sensitive techniques for the fast and consistent detection and quantification of target sets of proteins across multiple samples. This is only partly achieved by classical mass spectrometry or affinity-based methods. We applied a targeted proteomics approach based on selected reaction monitoring (SRM) to detect and quantify proteins expressed to a concentration below 50 copies/cell in total S. cerevisiae digests. The detection range can be extended to single-digit copies/cell and to proteins undetected by classical methods. We illustrate the power of the technique by the consistent and fast measurement of a network of proteins spanning the entire abundance range over a growth time course of S. cerevisiae transiting through a series of metabolic phases. We therefore demonstrate the potential of SRM-based proteomics to provide assays for the measurement of any set of proteins of interest in yeast at high-throughput and quantitative accuracy.

Intersection of ChIP and FLIP, genomic methods to study the dynamics of the cohesin proteins

The evolutionarily conserved cohesin proteins Smc1, Smc3, Rad21 (Mcd1), and Scc3 function in the cohesin complex that provides the basis for chromosome cohesion and is involved in gene regulation. Understanding how these proteins link together the genome requires the use of whole-genome approaches to study the molecular mechanisms of these essential proteins. While chromatin immunoprecipitation followed by DNA microarray (ChIP-chip) studies have provided a snapshot in time of where these proteins associate with various genomes, the cohesin proteins are dynamic in their localization and interactions on chromatin. Study of the dynamic nature of these proteins requires approaches such as live cell imaging. We present evidence from fluorescence loss in photobleaching (FLIP) experiments in budding yeast that the decay constant of each cohesin subunit is approximately 60-90 s in interphase. The decay constant on chromatin increases from G(1) to S phase to metaphase, consistent with the interaction with chromatin becoming more stable once chromosomes are cohered. A small population of Smc3 at a position consistent with centromeric location has a longer decay constant than bulk Smc3. The characterization of the interaction of cohesin with chromatin, in terms of both its position and its dynamics, may be key to understanding how this protein complex contributes to chromosome segregation and gene regulation.

Comparative proteomic analysis of tolerance and adaptation of ethanologenic Saccharomyces cerevisiae to furfural, a lignocellulosic inhibitory compound

The molecular mechanism involved in tolerance and adaptation of ethanologenic Saccharomyces cerevisiae to inhibitors (such as furfural, acetic acid, and phenol) represented in lignocellulosic hydrolysate is still unclear. Here, (18)O-labeling-aided shotgun comparative proteome analysis was applied to study the global protein expression profiles of S. cerevisiae under conditions of treatment of furfural compared with furfural-free fermentation profiles. Proteins involved in glucose fermentation and/or the tricarboxylic acid cycle were upregulated in cells treated with furfural compared with the control cells, while proteins involved in glycerol biosynthesis were downregulated. Differential levels of expression of alcohol dehydrogenases were observed. On the other hand, the levels of NADH, NAD(+), and NADH/NAD(+) were reduced whereas the levels of ATP and ADP were increased. These observations indicate that central carbon metabolism, levels of alcohol dehydrogenases, and the redox balance may be related to tolerance of ethanologenic yeast for and adaptation to furfural. Furthermore, proteins involved in stress response, including the unfolded protein response, oxidative stress, osmotic and salt stress, DNA damage and nutrient starvation, were differentially expressed, a finding that was validated by quantitative real-time reverse transcription-PCR to further confirm that the general stress responses are essential for cellular defense against furfural. These insights into the response of yeast to the presence of furfural will benefit the design and development of inhibitor-tolerant ethanologenic yeast by metabolic engineering or synthetic biology.

PTC1 is required for vacuole inheritance and promotes the association of the myosin-V vacuole-specific receptor complex

Organelle inheritance occurs during cell division. In Saccharomyces cerevisiae, inheritance of the vacuole, and the distribution of mitochondria and cortical endoplasmic reticulum are regulated by Ptc1p, a type 2C protein phosphatase. Here we show that PTC1/VAC10 controls the distribution of additional cargoes moved by a myosin-V motor. These include peroxisomes, secretory vesicles, cargoes of Myo2p, and ASH1 mRNA, a cargo of Myo4p. We find that Ptc1p is required for the proper distribution of both Myo2p and Myo4p. Surprisingly, PTC1 is also required to maintain the steady-state levels of organelle-specific receptors, including Vac17p, Inp2p, and Mmr1p, which attach Myo2p to the vacuole, peroxisomes, and mitochondria, respectively. Furthermore, Vac17p fused to the cargo-binding domain of Myo2p suppressed the vacuole inheritance defect in ptc1Delta cells. These findings suggest that PTC1 promotes the association of myosin-V with its organelle-specific adaptor proteins. Moreover, these observations suggest that despite the existence of organelle-specific receptors, there is a higher order regulation that coordinates the movement of diverse cellular components.

Radon-222 exhalation from open ground on and around a uranium mine in the wet-dry tropics

Radon-222 exhalation from the ground surface depends upon a number of variables such as the 226Ra activity concentration and its distribution in soil grains; soil grain size; soil porosity, temperature and moisture; atmospheric pressure, rainfall and temperature. In this study, 222Rn exhalation flux density measurements within and around the Ranger uranium mine in northern Australia were performed to investigate the effect of these variables within a tropical region. Measurements were taken at the waste rock dumps, ore stockpiles, mine pits, and at sites where effluent water with elevated 226Ra concentration has been spray irrigated over land, as well as at sites outside the mine. The sites selected represented a variety of geomorphic regions ranging from uranium-bearing rocks to ambient soils. Generally, wet season rains reduced 222Rn exhalation but at a few sites the onset of rains caused a step rise in exhalation flux densities. The results show that parameters such as 226Ra activity concentration, soil grain size and soil porosity have a marked effect on 222Rn flux densities. For similar geomorphic sites, 226Ra activity concentration is a dominant factor, but soil grain size and porosity also influence 222Rn exhalation. Surfaces with vegetation showed higher exhalation flux densities than their barren counterparts, perhaps because the associated root structure increases soil porosity and moisture retention. Repeated measurements over one year at eight sites enabled an analysis of precipitation and soil moisture effects on 222Rn exhalation. Soil moisture depth profiles varied both between seasons and at different times during the wet season, indicating that factors such as duration, intensity and time between precipitation events can influence 222Rn flux densities considerably.

Capillary isoelectric focusing/reversed phase liquid chromatography/mass spectrometry

The vast number of proteins present in the proteome of a typical organism requires that separations be performed on the mixture prior to introduction into a mass spectrometer for protein identification and quantification. An integrated protein separation platform, combining capillary isoelectric focusing (CIEF) with reversed phase liquid chromatography (RPLC), is described to provide high resolving power for the analysis of complex protein mixtures. Thus, the proteins are systematically resolved according to their differences in isoelectric point and hydrophobicity using combined CIEF/RPLC separations. A key feature of the CIEF-based multidimensional separation platform is the elimination of protein loss and dilution in an integrated platform while achieving comprehensive and ultrasensitive analysis of protein profiles within small cell populations or limited tissue samples.

Mining protein networks for synthetic genetic interactions

BACKGROUND

The local connectivity and global position of a protein in a protein interaction network are known to correlate with some of its functional properties, including its essentiality or dispensability. It is therefore of interest to extend this observation and examine whether network properties of two proteins considered simultaneously can determine their joint dispensability, i.e., their propensity for synthetic sick/lethal interaction. Accordingly, we examine the predictive power of protein interaction networks for synthetic genetic interaction in Saccharomyces cerevisiae, an organism in which high confidence protein interaction networks are available and synthetic sick/lethal gene pairs have been extensively identified.

RESULTS

We design a support vector machine system that uses graph-theoretic properties of two proteins in a protein interaction network as input features for prediction of synthetic sick/lethal interactions. The system is trained on interacting and non-interacting gene pairs culled from large scale genetic screens as well as literature-curated data. We find that the method is capable of predicting synthetic genetic interactions with sensitivity and specificity both exceeding 85%. We further find that the prediction performance is reasonably robust with respect to errors in the protein interaction network and with respect to changes in the features of test datasets. Using the prediction system, we carried out novel predictions of synthetic sick/lethal gene pairs at a genome-wide scale. These pairs appear to have functional properties that are similar to those that characterize the known synthetic lethal gene pairs.

CONCLUSION

Our analysis shows that protein interaction networks can be used to predict synthetic lethal interactions with accuracies on par with or exceeding that of other computational methods that use a variety of input features, including functional annotations. This indicates that protein interaction networks could plausibly be rich sources of information about epistatic effects among genes.