Search

With the current COVID-19 pandemic, we all realized how important interactions are. Interactions are everywhere. At the cellular level, protein interactions play a key role and their ensemble, also called interactome, is often referred as the basic building blocks of life. Given its importance, the maintenance of the integrity of the interactome is a real challenge in the cell. Many events during evolution can disrupt interactomes and potentially result in different characteristics for the organisms. However, the molecular underpinnings of changes in interactions at the cellular level are still largely unexplored. Among the perturbations, hybridization puts in contact two different interactomes, which may lead to many changes in the protein interaction network of the hybrid, including gains and losses of interactions. We recently investigated the fate of the interactomes after hybridization between yeast species using a comparative proteomics approach. A large-scale conservation of the interactions was observed in hybrids, but we also noticed the presence of proteostasis-related changes. This suggests that, despite a general robustness, small differences may accumulate in hybrids and perturb their protein physiology. Here, we summarize our work with a broader perspective on the importance of interactions.

The possibility to perform bioprocesses with reduced ecological footprint to produce natural compounds and catalyzers of industrial interest is pushing the research for salt tolerant microorganisms able to grow on seawater-based media and able to use a wide range of nutrients coming from waste. In this study we focused our attention on a Debaryomyces hansenii marine strain (Mo40). We optimized cultivation in a bioreactor at low pH on seawater-based media containing a mixture of sugars (glucose and xylose) and urea. Under these conditions the strain exhibited high growth rate and biomass yield. In addition, we characterized potential applications of this yeast biomass in food/feed industry. We show that Mo40 can produce a biomass containing 45% proteins and 20% lipids. This strain is also able to degrade phytic acid by a cell-bound phytase activity. These features represent an appealing starting point for obtaining D. hansenii biomass in a cheap and environmentally friendly way, and for potential use as an additive or to replace unsustainable ingredients in the feed or food industries, as this species is included in the QPS EFSA list (Quality Presumption as Safe—European Food Safety Authority).

Nearly all cellular functions are mediated by protein-protein interactions and mapping the interactome provides fundamental insights into the regulation and structure of biological systems. In principle, affinity purification coupled to mass spectrometry (AP-MS) is an ideal and scalable tool, however, it has been difficult to identify low copy number complexes, membrane complexes and those disturbed by protein-tagging. As a result, our current knowledge of the interactome is far from complete, and assessing the reliability of reported interactions is challenging. Here we develop a sensitive, high-throughput, and highly reproducible AP-MS technology combined with a quantitative two-dimensional analysis strategy for comprehensive interactome mapping of Saccharomyces cerevisiae. We reduced required cell culture volumes thousand-fold and employed 96-well formats throughout, allowing replicate analysis of the endogenous green fluorescent protein (GFP) tagged library covering the entire expressed yeast proteome. The 4159 pull-downs generated a highly structured network of 3,909 proteins connected by 29,710 interactions. Compared to previous large-scale studies, we double the number of proteins (nodes in the network) and triple the number of reliable interactions (edges), including very low abundant epigenetic complexes, organellar membrane complexes and non-taggable complexes interfered by abundance correlation. This nearly saturated interactome reveals that the vast majority of yeast proteins are highly connected, with an average of 15 interactors, the majority of them unreported so far. Similar to social networks between humans, the average shortest distance is 4.2 interactions. A web portal (www.yeast-interactome.org) enables exploration of our dataset by the network and biological communities and variations of our AP-MS technology can be employed in any organism or dynamic conditions.

Summary Synthetic proteins with high affinity and selectivity for a protein target can be used as research tools, biomarkers, and pharmacological agents, but few methods exist to design such proteins de novo. To this end, the In-Silico Protein Synthesizer (InSiPS) was developed to design synthetic binding proteins (SBPs) that bind pre-determined targets while minimizing off-target interactions. InSiPS is a genetic algorithm that refines a pool of random sequences over hundreds of generations of mutation and selection to produce SBPs with pre-specified binding characteristics. As a proof of concept, we design SBPs against three yeast proteins and demonstrate binding and functional inhibition of two of three targets in vivo. Peptide SPOT arrays confirm binding sites, and a permutation array demonstrates target specificity. Our foundational approach will support the field of de novo design of small binding polypeptide motifs and has robust applicability while offering potential advantages over the limited number of techniques currently available., Graphical Abstract, Highlights • InSiPS engineers synthetic binding proteins (SBPs) using primary protein sequence • SBPs are designed to a bind a target protein and avoid “off-target” interactions • Binding and functional inhibition of two of three target proteins in yeast is demonstrated • Our new approach offers advantages over alternative tools that rely on 3D models, Biological Sciences; Bioinformatics; Protein Family Determination

More than a hundred proteins in yeast reversibly aggregate and phase-separate in response to various stressors, such as nutrient depletion and heat shock. We know little about the protein sequence and structural features behind this ability, which has not been characterized on a proteome-wide level. To identify the distinctive features of aggregation-prone protein regions, we apply machine learning algorithms to genome-scale limited proteolysis-mass spectrometry (LiP-MS) data from yeast proteins. LiP-MS data reveals that 96 proteins show significant structural changes upon heat shock. We find that in these proteins the propensity to phase separate cannot be solely driven by disordered regions, because their aggregation-prone regions (APRs) are not significantly disordered. Instead, the phase separation of these proteins requires contributions from both disordered and structured regions. APRs are significantly enriched in aliphatic residues and depleted in positively charged amino acids. Aggregator proteins with longer APRs show a greater propensity to aggregate, a relationship that can be explained by equilibrium statistical thermodynamics. Altogether, our observations suggest that proteome-wide reversible protein aggregation is mediated by sequence-encoded properties. We propose that aggregating proteins resemble supra-molecular amphiphiles, where APRs are the hydrophobic parts, and non-APRs are the hydrophilic parts.

More than a hundred proteins in yeast reversibly aggregate and phase-separate in response to various stressors, such as nutrient depletion and heat shock. We know little about the sequence and structural features behind this ability, which has not been characterized on a proteome-wide level. To identify the distinctive features of aggregation-prone regions, we apply machine learning algorithms to genome-scale limited proteolysis-mass spectrometry data from 96 yeast proteins that phase-separate upon heat shock. We find that the propensity to phase separate cannot be solely driven by disordered regions, because the aggregation-prone regions of our study proteins are not significantly disordered. Instead, the phase separation of these proteins requires contributions from both disordered and structured regions. APRs are significantly enriched in aliphatic residues and depleted in positively charged amino acids. Aggregator proteins with longer APRs show a greater propensity to aggregate, a relationship that can be explained by equilibrium statistical thermodynamics. Altogether, our observations suggest that proteome-wide reversible protein aggregation is mediated by sequence-encoded properties. Aggregating proteins resemble supra-molecular amphiphiles, where APRs are the hydrophobic parts, and non-APRs are the hydrophilic parts.

Background Skin prick test (SPT) solutions and allergy vaccines (AVs) are crucial tools for diagnosis and therapy of allergies. It was the aim of this study to corroborate the content of products for diagnosis and treatment of dust mite allergies that are produced and sold in India. Methods SDS-PAGE, immunoblots and high-resolution mass spectrometric analysis was performed with 16 house dust mite (HDM) SPT solutions and AVs from 3 Indian manufacturers. Authority-approved European SPT solutions and in-house extracts were used as references. Results From the 5 Indian Dermatophagoides pteronyssinus products, none contained proteins from this source. Instead, 1 sample contained Dermatophagoides farinae and human serum proteins, 4 products contained allergens from the storage mite Suidasia medanensis, allergens from the legume Cicer arietinum (chickpea), and proteins from baker's yeast. From 4 Indian D. farinae-labeled products, 2 contained human serum proteins and a limited number of D. farinae allergens. Two contained only Suidasia, Cicer, and yeast proteins. In contrast, the European authority-approved D. pteronyssinus and D. farinae SPT solutions that were used as reference in this study, contained exclusively proteins of the respective species and covered the expected allergen spectra. The Blomia tropicalis sample contained no Blomia allergens at all, but consisted exclusively of Suidasia, Cicer, and yeast proteins. All 6 HDM samples consisted of human serum proteins and limited amounts of D. farinae allergens. Conclusions All commercial Indian SPT solutions and AVs analyzed in this study are not suitable for dust mite allergy diagnosis and therapy, as they contain either no, or only a limited number of, HDM allergens. In addition, their use could lead to misdiagnosis since some of them contain allergens from other sources, including the storage mite Suidasia, chickpea, as well as baker's yeast. Further, their application might be harmful to patients, as some products contain large amounts of proteins of human origin. Analysis of European SPT solutions, on the other hand, confirmed their suitability for dust mite allergy diagnosis.

Proteins obtained from microbial biomass like bacteria, yeast, and algae, coined as single cell proteins (SCP), are considered as new and unconventional sources of protein supply both for animal and human use. These microbes convert low value by-products and wastes into protein rich biomass through the process of fermentation. In this regard, yeast was the first microorganism used as source of protein in animal feed. Different industrial methods including plasmolysis, autolysis or enzyme digestions are used to extract the protein rich inner content of a yeast cell known as yeast extract (YE). The protein content of YE may vary depending on the method of extraction, but generally ranges between 50–60% and contains free amino acids and functional peptides. In swine the weaning transition is a critical period where piglets are exposed to different kinds of stresses including the abrupt change from milk to solid feed. To match the highly digestible and protein rich diet in milk, pig producers are using a highly complex diet containing animal products including spray-dried porcine plasma (SDPP) in the pre-starter diet. However, the use of animal products is under scrutiny for biosecurity reasons and SCP from yeast are emerging as good alternatives to replace animal products in piglet diets. Reports from various studies have shown proteins from YE can provide highly digestible proteins with good profile amino acids. Furthermore, it has been shown that yeast proteins improve health and performance of pigs by enhancing immunity and strengthening gut integrity of weaning piglets. Moreover, comparative animal studies in weaned piglets confirmed no significant difference between SDPP and YE proteins in energy content, AA composition, as well as apparent and standardized ileal digestibility of AA suggesting that SCP from yeast could be a good alternative to replace SDPP in weanling piglet diets.

The maintenance of duplicated genes is thought to protect cells from genetic perturbations, but the molecular basis of this robustness is largely unknown. By measuring the interaction of yeast proteins with their partners in wild-type cells and in cells lacking a paralog, we found that 22 out of 56 paralog pairs compensate for the lost interactions. An equivalent number of pairs exhibit the opposite behavior and require each other's presence for maintaining their interactions. These dependent paralogs generally interact physically, regulate each other's abundance, and derive from ancestral self-interacting proteins. This reveals that gene duplication may actually increase mutational fragility instead of robustness in a large number of cases.

Stochasticity between independent LC–MS/MS runs is a challenging problem in the field of proteomics, resulting in significant missing values (i.e., abundance measurements) among observed peptides. To address this issue, several approaches have been developed including computational methods such as MaxQuant’s match-between-runs (MBR) algorithm. Often dozens of runs are all considered at once by MBR, transferring identifications from any one run to any of the others. To evaluate the error associated with these transfer events, we created a two-sample/two-proteome approach. In this way, samples containing no yeast lysate (n = 20) were assessed for false identification transfers from samples containing yeast (n = 20). While MBR increased the total number of spectral identifications by ~40%, we also found that 44% of all identified yeast proteins had identifications transferred to at least one sample without yeast. However, of these only 2.7% remained in the final data set after applying the MaxQuant LFQ algorithm. We conclude that false transfers by MBR are plentiful,0020but few are retained in the final data set.

Heat-stress triggers the formation of condensates known as stress granules (SGs), which store non-translating mRNA and stalled translation initiation complexes. To gain a better understanding of SGs, we identified yeast proteins that sediment after heat-shock by mass spectrometry. Heat-regulated proteins are biased toward a subset of abundant proteins that are significantly enriched in intrinsically disordered regions (IDRs). SG localization of over 80 heat-regulated proteins was confirmed using microscopy, including 32 proteins that were not known previously to localize to SGs. We find that several IDRs are sufficient to mediate SG recruitment. Moreover, the diffusive exchange of IDRs within SGs, observed via FRAP, can be highly dynamic while other components remain immobile. Lastly, we showed that the IDR of the Ubp3 deubiquitinase is critical for SG formation. This work confirms that IDRs play an important role in cellular compartmentalization upon stress, can be sufficient for SG incorporation, can remain dynamic in vitrified SGs, and play a vital role during heat-stress.SummaryThe authors provide an in-depth proteomic study of yeast heat stress granule (SG) proteins. They identified intrinsic disordered regions (IDRs) as one of the main features shared by these proteins and demonstrated IDRs can be sufficient for SG recruitment.

The knowledge of subcellular localization of proteins can provide useful clues about their functions. The conventional methods to determine the subcellular localization are unable to keep pace with the rate at which the new data is being generated. Thus, though sequence information is available, the localization and function of a number of proteins remains unknown. In this study, we have developed a script that makes use of the physical interactors of a protein and their localization data to predict the subcellular localization. We used the script to predict the localization of yeast proteins for which there is no localization data. Further, we experimentally verified the predicted localization for six arbitrarily chosen proteins and found our predictions to be correct for five of the proteins.

The N-end rule denotes the relationship between the identity of the amino-terminal residue of a protein and its in vivo half-life. Since its discovery in 1986, the N-end rule has generally been described by a defined set of rules for determining whether an amino-terminal residue is stabilizing or not. However, recent studies are revealing that this N-end rule (or N-degron concept) is less straightforward than previously appreciated. For instance, it is unveiled that N-terminal acetylation of N-terminal residues may create a degradation signal (Ac-degron) that promotes the degradation of target proteins. A recent high-throughput dissection of degrons in yeast proteins amino termini intriguingly suggested that the hydrophobicity of amino-terminal residues-but not the N-terminal acetylation status-may be the indispensable feature of amino-terminal degrons. Herein, these recent advances in N-terminal acetylation and the complexity of N-terminal degradation signals in the context of the N-degron pathway are analyzed.

The proper pausing of replication forks at barriers on chromosomes is important for genome integrity. However, the detailed mechanism underlying this process has not been well elucidated. Here, we successfully reconstituted fork-pausing reactions from purified yeast proteins on templates that had binding sites for the LacI, LexA, and/or Fob1 proteins; the forks paused specifically at the protein-bound sites. Moreover, although the replicative helicase Cdc45–Mcm2–7–GINS (CMG) complex alone unwound the protein-bound templates, the unwinding of the LacI-bound site was impeded by the presence of a main leading strand DNA polymerase: polymerase ε (Polε). This suggests that Polε modulates CMG to pause at these sites.

The ability to measure the abundance and visualize the localization of proteins across the yeast proteome has stimulated hypotheses on gene function and fueled discoveries. While the classic C’ tagged GFP yeast library has been the only resource for over a decade, the recent development of the SWAT technology has led to the creation of multiple novel yeast libraries where new-generation fluorescent reporters are fused at the N’ and C’ of open reading frames. Efficient access to these data requires a user interface to visualize and compare protein abundance, localization and co-localization across cells, strains, and libraries. YeastRGB (www.yeastRGB.org) was designed to address such a need, through a user-friendly interface that maximizes informative content. It employs a compact display where cells are cropped and tiled together into a ‘cell-grid.’ This representation enables viewing dozens of cells for a particular strain within a display unit, and up to 30 display units can be arrayed on a standard high-definition screen. Additionally, the display unit allows users to control zoom-level and overlay of images acquired using different color channels. Thus, YeastRGB makes comparing abundance and localization efficient, across thousands of cells from different strains and libraries.

Modern beer production is a complex industrial process. However, some of its biochemical details remain unclear. Using mass spectrometry proteomics, we have performed a global untargeted analysis of the proteins present across time during nano-scale beer production. Samples included sweet wort produced by a high temperature infusion mash, hopped wort, and bright beer. This analysis identified over 200 unique proteins from barley and yeast, emphasizing the complexity of the process and product. We then used data independent SWATH-MS to quantitatively compare the relative abundance of these proteins throughout the process. This identified large and significant changes in the proteome at each process step. These changes described enrichment of proteins by their biophysical properties, and identified the appearance of dominant yeast proteins during fermentation. Altered levels of malt modification also quantitatively changed the proteomes throughout the process. Detailed inspection of the proteomic data revealed that many proteins were modified by protease digestion, glycation, or oxidation during the processing steps. This work demonstrates the opportunities offered by modern mass spectrometry proteomics in understanding the ancient process of beer production.

Protein interaction networks are known to exhibit remarkable structures: scale-free and small-world and modular structures. To explain the evolutionary processes of protein interaction networks possessing scale-free and small-world structures, preferential attachment and duplication-divergence models have been proposed as mathematical models. Protein interaction networks are also known to exhibit another remarkable structural characteristic, modular structure. How the protein interaction networks became to exhibit modularity in their evolution? Here, we propose a hypothesis of modularity in the evolution of yeast protein interaction network based on molecular evolutionary evidence. We assigned yeast proteins into six evolutionary ages by constructing a phylogenetic profile. We found that all the almost half of hub proteins are evolutionarily new. Examining the evolutionary processes of protein complexes, functional modules and topological modules, we also found that member proteins of these modules tend to appear in one or two evolutionary ages. Moreover, proteins in protein complexes and topological modules show significantly low evolutionary rates than those not in these modules. Our results suggest a hypothesis of modularity in the evolution of yeast protein interaction network as systems evolution.

Green fluorescent protein (GFP) has become an invaluable tool in biological research. Many GFP variants have been created that differ in brightness, photostability, and folding robustness. We have created two hybrid GFP variants, Envy and Ivy, which we placed in a vector for the C-terminal tagging of yeast proteins by PCR-mediated recombination. The Envy GFP variant combines mutations found in the robustly folding SuperfolderGFP and GFPγ, while the Ivy GFP variant is a hybrid of GFPγ and the yellow-green GFP variant, Clover. We compared Envy and Ivy to EGFP, SuperfolderGFP and GFPγ and found that Envy is brighter than the other GFP variants at both 30°C and 37°C, while Ivy is the most photostable. Envy and Ivy are recognized by a commonly used anti-GFP antibody, and both variants can be immunoprecipitated using the GFP TRAP Camelidae antibody nanotrap technology. Because Envy is brighter than the other GFP variants and is as photostable as GFPγ, we suggest that Envy should be the preferred GFP variant, while Ivy may be used in cases where photostability is of the utmost importance.

This method was generated to isolate high affinity protein complexes from yeast lysate by performing serial affinity purification of doubly tagged 3×FLAG/V5 proteins. First, the bait protein of interest is immunoprecipitated by anti-FLAG-conjugated magnetic beads and gently eluted by 3×FLAG antigen peptide. Next, the bait protein is recaptured from the first eluate by anti-V5-conjugated magnetic beads and eluted with ionic detergent. In this manner, the majority of abundant, nonspecific proteins remain either bound to the first beads or in the first eluate, allowing pure, high affinity complexes to be obtained. This approach can be used to show specific interactions with notoriously ‘sticky’ chaperone proteins.

Wine producers have to label the use of certain animal proteins for fining of must and wine since 2012 [regulation (EC) No. 579/2012] under specific conditions, which has encouraged the search for alternative fining agents produced out of plants or yeasts. Yeast proteins are naturally present in wine and first studies of yeast protein extracts (YPE) for the fining of wine have been published, which were mainly focused on the technical properties of the YPE as fining agents. The present study used YPE of different production techniques for fining of red wine and evaluated their technical properties. Furthermore, the sensory effect of the new fining agents, their protein composition and the presence of glycans and glycoproteins were examined. The YPE showed differences in their protein profile and in protein glycosylation, but all YPE that were investigated were successfully applied for fining of red wine. Mannoproteins were detected in all YPE, and they did not prevent the mutual precipitation of proteins and tannins that is necessary for successful fining as reported earlier in case of glycans and glycoproteins. The chemical parameters of the red wine were modified by fining with YPE in a way comparable to fining with gelatine, a traditional fining agent used as reference, and the initial quality of wine flavour was maintained during the treatments.

Maya Schuldiner pursued her PhD degree under the guidance of Prof. Nissim Benvenisty at the Hebrew University in Jerusalem. She carried out her postdoctoral training in the laboratory of Prof. Jonathan Weissman at the University of California, San Francisco, with support from the Human Frontiers Science Program and the Sandler Fellows Program. Since 2008, Maya has been running her own laboratory at the Weizmann Institute of Sciences in Rehovot, Israel. She received the Human Frontiers Science Program Career Development Award and the NIH Pathway to Independence Award, and she is a member of the EMBO Young Investigator programme and Faculty of 1000. Her current research interests are focused on unravelling novel functions of yeast proteins that are involved in organelle biology.

The identification of protein–protein interactions (PPIs) and their networks is vitally important to systemically define and understand the roles of proteins in biological systems. In spite of development of numerous experimental systems to detect PPIs and diverse research on assessment of the quality of the obtained data, a consensus – highly reliable, almost complete – interactome of Saccharomyces cerevisiae is not presented yet. In this work, we proposed an unsupervised statistical approach to create a high-confidence yeast PPI network. For this, we assembled databases of interacting protein pairs for yeast and obtained an extremely large PPI dataset which comprises of 135 154 non-redundant interactions between 6191 yeast proteins. A scoring scheme considering eight heterogeneous biological features resulted with a broad score distribution and a highly reliable network consisting of 29 046 physical interactions with scores higher than the threshold value of 0.85, for which sensitivity, specificity and coverage were 86%, 68%, and 72%, respectively. We evaluated our method by comparing it with other scoring schemes and showed that reducing the noise inherent in experimental PPIs via our scoring scheme further increased the accuracy. Current study is expected to increase the efficiency of the methodologies in biological research which make use of protein interaction networks.

The purpose of this work was to determine the levels of protein and the amino acid distribution in the cell mass of yeast strains (Saccharomyces sensu stricto) originated from Brazilian bioethanol industries. The protein was analyzed with the Kjeldahl method and the amino acids, by using high-performance liquid chromatography (HPLC). The percentages of the protein found ranged from 39 to 49%. The results show that in spite of some variation in numbers between the different yeast strains, all of them presented an amino acid profile similar to the one in the literature for S. cerevisae. The amino acids that have occurred in the largest amounts were: aspartic, glutamic acids and lysine, and those in the lowest amounts were: cysteine and methionine. Although the characteristics of the feedstock used and the process conditions are determinant of the protein values obtained in dry mass, this work elucidates that the intrinsic properties of the yeast strain influence these values.

Protein microarrays have emerged as a powerful tool for the scientific community, and their greatest advantage lies in the fact that thousands of reactions can be performed in a parallel and unbiased manner. The first high-density protein microarray, dubbed the “yeast proteome array,” consisted of approximately 5800 full-length yeast proteins and was initially used to identify protein–lipid interactions. Further assays were subsequently developed to allow measurement of protein–DNA, protein–RNA, and protein–protein interactions, as well as four well-known posttranslational modifications: phosphorylation, acetylation, ubiquitylation, and SUMOylation. In this introduction, we describe the advent of high-density protein microarrays, as well as current methods for assessing a wide variety of protein interactions and posttranslational modifications.

Vegemite is an iconic Australian food spread made from spent brewers’ yeast extract, which has been reported to be used as an ingredient in illegal home brewing. In this study, we tested the utility of Vegemite and the similar spread Marmite in promoting fermentation. We could not culture microorganisms from either Vegemite or Marmite, consistent with these food-grade spreads being essentially sterile. To test if the addition of Vegemite or Marmite could assist in fermentation when additional viable yeast was also present, solutions containing glucose and a range of concentrations of either Vegemite or Marmite were inoculated with brewers’ yeast. No fermentation occurred in any condition without addition of extra brewer’s yeast. Fermentation did not occur when yeast was inoculated into solutions containing only glucose, but progressed efficiently with when Vegemite or Marmite was also added. Gas Chromatography confirmed that ethanol was present at ∼3% v/v post-fermentation in all samples which contained glucose, Vegemite or Marmite, and brewers’ yeast. Trace amounts of methanol were also detected. Mass spectrometry proteomics identified abundant intracellular yeast proteins and barley proteins in Vegemite and Marmite, and abundant secreted yeast proteins from actively growing yeast in those samples to which extra brewers’ yeast had been added. We estimate that the real-world cost of home brewed “Vegemite Beer” would be very low. Our results show that Vegemite or other yeast extract spreads could provide cheap and readily available sources of nutrient supplementation to increase the efficiency of fermentation in home brewing or other settings.

High throughput microscopy of many single cells generates high-dimensional data that are far from straightforward to analyze. One important problem is automatically detecting the cellular compartment where a fluorescently tagged protein resides, a task relatively simple for an experienced human, but difficult to automate on a computer. Here, we train an 11-layer neural network on data from mapping thousands of yeast proteins, achieving per cell localization classification accuracy of 91%, and per protein accuracy of 99% on held out images. We confirm that low-level network features correspond to basic image characteristics, while deeper layers separate localization classes. Using this network as a feature calculator, we train standard classifiers that assign proteins to previously unseen compartments after observing only a small number of training examples. Our results are the most accurate subcellular localization classifications to date, and demonstrate the usefulness of deep learning for high throughput microscopy.

Defining intracellular protein concentration is critical in molecular systems biology. Although strategies for determining relative protein changes are available, defining robust absolute values in copies per cell has proven significantly more challenging. Here we present a reference data set quantifying over 1800 Saccharomyces cerevisiae proteins by direct means using protein-specific stable-isotope labeled internal standards and selected reaction monitoring (SRM) mass spectrometry, far exceeding any previous study. This was achieved by careful design of over 100 QconCAT recombinant proteins as standards, defining 1167 proteins in terms of copies per cell and upper limits on a further 668, with robust CVs routinely less than 20%. The selected reaction monitoring-derived proteome is compared with existing quantitative data sets, highlighting the disparities between methodologies. Coupled with a quantification of the transcriptome by RNA-seq taken from the same cells, these data support revised estimates of several fundamental molecular parameters: a total protein count of ∼100 million molecules-per-cell, a median of ∼1000 proteins-per-transcript, and a linear model of protein translation explaining 70% of the variance in translation rate. This work contributes a “gold-standard” reference yeast proteome (including 532 values based on high quality, dual peptide quantification) that can be widely used in systems models and for other comparative studies. Reliable and accurate quantification of the proteins present in a cell or tissue remains a major challenge for post-genome scientists. Proteins are the primary functional molecules in biological systems and knowledge of their abundance and dynamics is an important prerequisite to a complete understanding of natural physiological processes, or dysfunction in disease. Accordingly, much effort has been spent in the development of reliable, accurate and sensitive techniques to quantify the cellular proteome, the complement of proteins expressed at a given time under defined conditions (1). Moreover, the ability to model a biological system and thus characterize it in kinetic terms, requires that protein concentrations be defined in absolute numbers (2, 3). Given the high demand for accurate quantitative proteome data sets, there has been a continual drive to develop methodology to accomplish this, typically using mass spectrometry (MS) as the analytical platform. Many recent studies have highlighted the capabilities of MS to provide good coverage of the proteome at high sensitivity often using yeast as a demonstrator system (4⇓⇓⇓⇓⇓–10), suggesting that quantitative proteomics has now “come of age” (1). However, given that MS is not inherently quantitative, most of the approaches produce relative quantitation and do not typically measure the absolute concentrations of individual molecular species by direct means. For the yeast proteome, epitope tagging studies using green fluorescent protein or tandem affinity purification tags provides an alternative to MS. Here, collections of modified strains are generated that incorporate a detectable, and therefore quantifiable, tag that supports immunoblotting or fluorescence techniques (11, 12). However, such strategies for copies per cell (cpc) quantification rely on genetic manipulation of the host organism and hence do not quantify endogenous, unmodified protein. Similarly, the tagging can alter protein levels - in some instances hindering protein expression completely (11). Even so, epitope tagging methods have been of value to the community, yielding high coverage quantitative data sets for the majority of the yeast proteome (11, 12). MS-based methods do not rely on such nonendogenous labels, and can reach genome-wide levels of coverage. Accurate estimation of absolute concentrations i.e. protein copy number per cell, also usually necessitates the use of (one or more) external or internal standards from which to derive absolute abundance (4). Examples include a comprehensive quantification of the Leptospira interrogans proteome that used a 19 protein subset quantified using selected reaction monitoring (SRM)1 to calibrate their label-free data (8, 13). It is worth noting that epitope tagging methods, although also absolute, rely on a very limited set of standards for the quantitative western blots and necessitate incorporation of a suitable immunogenic tag (11). Other recent, innovative approaches exploiting total ion signal and internal scaling to estimate protein cellular abundance (10, 14), avoid the use of internal standards, though they do rely on targeted proteomic data to validate their approach. The use of targeted SRM strategies to derive proteomic calibration standards highlights its advantages in comparison to label-free in terms of accuracy, precision, dynamic range and limit of detection and has gained currency for its reliability and sensitivity (3, 15⇓–17). Indeed, SRM is often referred to as the “gold standard proteomic quantification method,” being particularly well-suited when the proteins to be quantified are known, when appropriate surrogate peptides for protein quantification can be selected a priori, and matched with stable isotope-labeled (SIL) standards (18⇓–20). In combination with SIL peptide standards that can be generated through a variety of means (3, 15), SRM can be used to quantify low copy number proteins, reaching down to ∼50 cpc in yeast (5). However, although SRM methodology has been used extensively for S. cerevisiae protein quantification by us and others (19, 21, 22), it has not been used for large protein cohorts because of the requirement to generate the large numbers of attendant SIL peptide standards; the largest published data set is only for a few tens of proteins. It remains a challenge therefore to robustly quantify an entire eukaryotic proteome in absolute terms by direct means using targeted MS and this is the focus of our present study, the Census Of the Proteome of Yeast (CoPY). We present here direct and absolute quantification of nearly 2000 endogenous proteins from S. cerevisiae grown in steady state in a chemostat culture, using the SRM-based QconCAT approach. Although arguably not quantification of the entire proteome, this represents an accurate and rigorous collection of direct yeast protein quantifications, providing a gold-standard data set of endogenous protein levels for future reference and comparative studies. The highly reproducible SIL-SRM MS data, with robust CVs typically less than 20%, is compared with other extant data sets that were obtained via alternative analytical strategies. We also report a matched high quality transcriptome from the same cells using RNA-seq, which supports additional calculations including a refined estimate of the total protein content in yeast cells, and a simple linear model of translation explaining 70% of the variance between RNA and protein levels in yeast chemostat cultures. These analyses confirm the validity of our data and approach, which we believe represents a state-of-the-art absolute quantification compendium of a significant proportion of a model eukaryotic proteome.

Proteomics is an emerging area of systems biology that allows simultaneous study of thousands of proteins expressed in cells, tissues, or whole organisms. We have developed this activity to enable high school or college students to explore proteomic databases using mass spectrometry data files generated from yeast proteins in a college laboratory course. Students upload files of “unknown” proteins from our public website, enter them into a proteomics search engine (Mascot), identify the proteins, and use additional proteomics websites to learn about their functions and three-dimensional structures. This activity is suitable for use in units exploring protein structure and function, metabolism, or bioinformatics.

Elvers and glass eels of Anguilla anguilla were fed diets containing two types of feeding stimulants (FS) that were based on processed marine (MBFS) and yeast proteins (YBFS). Elvers (1.5 ± 0.3 g) were fed seven diets (MBFS and YBFS diets at 20 g kg−1, 40 g kg−1 and 60 g kg−1 plus control) for 60 days. Glass eels (250 ± 100 mg) were weaned to 60 g kg−1 MBFS, 60 g kg−1 YBFS and control diets for 30 days. Diets containing 60 g kg−1 FS had a beneficial effect in terms of growth, homogenous size distribution and feed intake in elvers. Elvers fed 60 g kg−1 MBFS and YBFS diets grew 11.9% and 5.6% faster than the control group. No differences in growth and size distribution were detected in glass eels fed 60 g kg−1 MBFS and YBFS diets. However, FS affected the digestive system maturation; fish fed the 60 g kg−1 MBFS and YBFS diets showed higher and intermediate values in the specific enzyme activities in comparison with the control group. This study revealed that the incorporation of FS into a pelleted diet was beneficial on the overall performance of European glass eels and elvers. However, the observed results were different depending on the eel’s stage of development, as well as the type and inclusion level of the FS.

A biphasic monolithic capillary column with 10 cm segment of strong-cation exchange monolith and 65 cm segment of reversed-phase monolith was prepared within a single 100 microm i.d. capillary. Separation performance of this column was evaluated by a five-cycle online multidimensional separation of 10 microg tryptic digest of yeast proteins using nanoflow liquid chromatography coupled with tandem mass spectrometry, and it took 12 h for whole separation under the operating pressure only approximately 900 psi. Totally, 780 distinct proteins were positively identified through assignment of 2953 unique peptides at false-positive rate less than 1%. The good separation performance of this biphasic column was largely attributed to the good orthogonality of the strong-cation exchange monolith and reversed-phase monolith for multidimensional separation.

Tfs1p and Ylr179cp are yeast proteins belonging to the PEBP family. Tfs1p, but not Ylr179cp, has been shown to interact with and inhibit Ira2p, a GTPase-activating protein of Ras. Tfs1p has been shown to be a specific inhibitor of the CPY protease and the 3D structure of the complex has been resolved. To shed light on the molecular determinants of Tfs1p involved in the Tfs1/Ira2 interaction, the 3D structure of Ylr179cp has been modelled and compared to that of Tfs1p. Tfs1p point mutants and Tfs1 hybrid proteins combining regions of Tfs1p and Ylr179cp were also designed and their function was tested. Results, interpreted from a structural point of view, show that the accessibility of the surface pocket of Tfs1p, its N-terminal region and the specific electrostatic properties of a large surface region containing these two elements, play a crucial role in this interaction.

This study was designed to characterize the suppressant effect of yeast protein and purified peptides on energy intake. For this purpose, 5 experiments were carried out using adult male Wistar rats. Rats that consumed ad libitum a standard yeast protein diet ate significantly less and were leaner over 21 d than rats that consumed ad libitum a standard milk protein diet (Expt. 1). Moreover, rats fed a high yeast protein load reduced their next meal and daily energy intake more than rats fed any other well-balanced, amino acid, high protein load (soy, total milk protein, or wheat gluten) and more than those fed a wheat starch diet (Expt. 2). Purified peptides from the yeast protein extract produced similar effects on subsequent energy intake (Expt. 3). Study of the behavioral satiety sequence showed that rats consuming P14-y or P55-y diets ad libitum did not acquire a conditioned food aversion (Expt. 4). Finally, a preliminary study of gastric emptying in rats fed yeast protein loads showed that yeast protein was emptied more rapidly through the pylorus than total milk protein during a meal, which may induce satiety (Expt. 5). Taken together, these experiments show that yeast proteins enhance satiety in rats more than other proteins.

Gel electrophoresis is a fractionation/separation technique that yields valuable information in the field of metalloproteomics, often referred to as metallomics. This paper is based on four years of practical experience of the authors' lab in this domain and highlights the capabilities and limitations of gel electrophoresis. Pitfalls of the technique were recognized by identifying the origin of artefacts in the separation, species degradation being the most important. Gel electrophoresis can be accomplished under either native or denaturing conditions. The speciation of vanadium and selenium among serum and yeast proteins, respectively, is used to illustrate these two major modes. The most powerful approach is two-dimensional denaturing gel electrophoresis. This review of the methods used in our laboratory also describes the application of the two major detection techniques, autoradiography on the one hand and electrothermal vaporization - or laser ablation (LA) - inductively coupled plasma-mass spectrometry (ICP-MS) on the other.

The product of the human Tre2 oncogene is structurally related to the Ypt/Rab GTPase-activating proteins (Ypt/Rab GAPs). Particularly, the oncoprotein shares with the yeast proteins Msb3p and Msb4p, and with the human protein RN-tre the highly conserved TBC domain, forming the catalytically active domain of Ypt/Rab GAPs. Yet, the Tre2 oncogene seems to encode a nonfunctional Rab GAP. As regions flanking the TBC domain may be crucial for catalytic activity, regions located N- and C-terminally with respect to this domain were explored. For this, chimeric proteins created by sequence exchanges between the Tre2 oncoprotein and RN-tre were tested for their ability to replace functionally the Msb3p and Msb4p proteins in double-mutant yeast cells. These complementation experiments revealed, in addition to the TBC domain, a second Tre2 region involved in the oncoprotein's lack of GAP activity: a 93-aa region flanking the TBC domain on the C-terminal side.

Ubiquitin-protein ligases (E3s) are implicated in various human disorders and are attractive targets for therapeutic intervention. Although most cellular proteins are ubiquitinated, ubiquitination cannot be linked directly to a specific E3 for a large fraction of these proteins, and the substrates of most E3 enzymes are unknown. We have developed a luminescent assay to detect ubiquitination in vitro, which is more quantitative, effective, and sensitive than conventional ubiquitination assays. By taking advantage of the abundance of purified proteins made available by genomic efforts, we screened hundreds of purified yeast proteins for ubiquitination, and we identified previously reported and novel substrates of the yeast E3 ligase Rsp5. The relevance of these substrates was confirmed in vivo by showing that a number of them interact genetically with Rsp5, and some were ubiquitinated by Rsp5 in vivo. The combination of this sensitive assay and the availability of purified substrates will enable the identification of substrates for any purified E3 enzyme.

This paper investigates the prospects of successful mass spectrometric protein identification based on mass data from proteolytic digests of complex protein mixtures. Sets of proteolytic peptide masses representing various numbers of digested proteins in a mixture were generated in silico. In each set, different proteins were selected from a protein sequence collection and for each protein the sequence coverage was randomly selected within a particular regime (15-30% or 30-60%). We demonstrate that the Probity algorithm, which is characterized by an optimal tolerance for random interference, employed in an iterative procedure can correctly identify95% of proteins at a desired significance level in mixtures composed of hundreds of yeast proteins under realistic mass spectrometric experimental constraints. By using a model of the distribution of protein abundance, we demonstrate that the very high efficiency of identification of protein mixtures that can be achieved by appropriate choices of informatics procedures is hampered by limitations of the mass spectrometric dynamic range. The results stress the desire to choose carefully experimental protocols for comprehensive proteome analysis, focusing on truly critical issues such as the dynamic range, which potentially limits the possibilities of identifying low abundance proteins.

Summary: The Human Protein Interaction Database (http://www.hpid.org) was designed (1) to provide human protein interaction information pre-computed from existing structural and experimental data, (2) to predict potential interactions between proteins submitted by users and (3) to provide a depository for new human protein interaction data from users. Two types of interaction are available from the pre-computed data: (1) interactions at the protein superfamily level and (2) those transferred from the interactions of yeast proteins. Interactions at the superfamily level were obtained by locating known structural interactions of the PDB in the SCOP domains and identifying homologs of the domains in the human proteins. Interactions transferred from yeast proteins were obtained by identifying homologs of the yeast proteins in the human proteins. For each human protein in the database and each query submitted by users, the protein superfamilies and yeast proteins assigned to the protein are shown, along with their interacting partners. We have also developed a set of web-based programs so that users can visualize and analyze protein interaction networks in order to explore the networks further. Availability: http://www.hpid.org Supplementary information: http://wilab.inha.ac.kr/HPID/Statistics.htm

Peptides are potentially useful for target validation and other reverse genetic applications. For instance, if a specific protein is susceptible to peptide inhibition, it may have a higher probability of being vulnerable to small molecules. We used the yeast two-hybrid technique to identify and study peptide binders for three yeast proteins involved in pheromone response: Ste11p, Ste18p, and Ste50p. A subset of peptide binders was shown to inhibit pheromone response in cells using two different functional assays. In addition, we utilized a variant of the yeast two-hybrid method to examine relative binding affinities based on competitive interactions in yeast. Our results suggest that binding affinity and inhibitory potency of peptides do not correlate perfectly and that peptide–protein interactions can be complex and unpredictable. Taken together these results suggest that while peptides are useful as in vivo inhibitors of protein function, caution must be exercised when choosing peptides for further studies and when inferring affinities from expression phenotypes.

In this postgenomic era, the ability to identify protein–protein interactions on a genomic scale is very important to assist in the assignment of physiological function. Because of the increasing number of solved structures involving protein complexes, the time is ripe to extend threading to the prediction of quaternary structure. In this spirit, a multimeric threading approach has been developed. The approach is comprised of two phases. In the first phase, traditional threading on a single chain is applied to generate a set of potential structures for the query sequences. In particular, we use our recently developed threading algorithm, PROSPECTOR. Then, for those proteins whose template structures are part of a known complex, we rethread on both partners in the complex and now include a protein–protein interfacial energy. To perform this analysis, a database of multimeric protein structures has been constructed, the necessary interfacial pairwise potentials have been derived, and a set of empirical indicators to identify true multimers based on the threading Z-score and the magnitude of the interfacial energy have been established. The algorithm has been tested on a benchmark set comprised of 40 homodimers, 15 heterodimers, and 69 monomers that were scanned against a protein library of 2478 structures that comprise a representative set of structures in the Protein Data Bank. Of these, the method correctly recognized and assigned 36 homodimers, 15 heterodimers, and 65 monomers. This protocol was applied to identify partners and assign quaternary structures of proteins found in the yeast database of interacting proteins. Our multimeric threading algorithm correctly predicts 144 interacting proteins, compared to the 56 (26) cases assigned by PSI-BLAST using a (less) permissive E-value of 1 (0.01). Next, all possible pairs of yeast proteins have been examined. Predictions (n = 2865) of protein–protein interactions are made; 1138 of these 2865 interactions have counterparts in the Database of Interacting Proteins. In contrast, PSI-BLAST made 1781 predictions, and 1215 have counterparts in DIP. An estimation of the false-negative rate for yeast-predicted interactions has also been provided. Thus, a promising approach to help assist in the assignment of protein–protein interactions on a genomic scale has been developed. Proteins 2002;49:350–364. © 2002 Wiley-Liss, Inc.

Protein aggregation is a widespread phenomenon in cells and associated with pathological conditions. Yet, little is known about the rules that govern protein aggregation in living cells. In this study, we biochemically isolated aggregation-prone proteins and used computational analyses to identify characteristics that are linked to physiological and arsenite-induced aggregation in living yeast cells. High protein abundance, extensive physical interactions, and certain structural properties are positively correlated with an increased aggregation propensity. The aggregated proteins have high translation rates and are substrates of ribosome-associated Hsp70 chaperones, indicating that they are susceptible for aggregation primarily during translation/folding. The aggregation-prone proteins are enriched for multiple chaperone interactions, thus high protein abundance is probably counterbalanced by molecular chaperones to allow soluble expression in vivo. Our data support the notion that arsenite interferes with chaperone activity and indicate that arsenite-aggregated proteins might engage in extensive aberrant protein–protein interactions. Expression of aggregation-prone proteins is down-regulated during arsenite stress, possibly to prevent their toxic accumulation. Several aggregation-prone yeast proteins have human homologues that are implicated in misfolding diseases, suggesting that similar mechanisms may apply in disease- and non-disease settings.

Yeast prions are heritable amyloid aggregates of functional yeast proteins; their propagation to subsequent cell generations is dependent upon fragmentation of prion protein aggregates by molecular chaperone proteins. Mounting evidence indicates the J-protein Sis1 may act as an amyloid specificity factor, recognizing prion and other amyloid aggregates and enabling Ssa and Hsp104 to act in prion fragmentation. Chaperone interactions with prions, however, can be affected by variations in amyloid-core structure resulting in distinct prion variants or ‘strains’. Our genetic analysis revealed that Sis1 domain requirements by distinct variants of [PSI +] are strongly dependent upon overall variant stability. Notably, multiple strong [PSI +] variants can be maintained by a minimal construct of Sis1 consisting of only the J-domain and glycine/phenylalanine-rich (G/F) region that was previously shown to be sufficient for cell viability and [RNQ +] prion propagation. In contrast, weak [PSI +] variants are lost under the same conditions but maintained by the expression of an Sis1 construct that lacks only the G/F region and cannot support [RNQ +] propagation, revealing mutually exclusive requirements for Sis1 function between these two prions. Prion loss is not due to [PSI +]-dependent toxicity or dependent upon a particular yeast genetic background. These observations necessitate that Sis1 must have at least two distinct functional roles that individual prions differentially require for propagation and which are localized to the glycine-rich domains of the Sis1. Based on these distinctions, Sis1 plasmid-shuffling in a [PSI +]/[RNQ +] strain permitted J-protein-dependent prion selection for either prion. We also found that, despite an initial report to the contrary, the human homolog of Sis1, Hdj1, is capable of [PSI +] prion propagation in place of Sis1. This conservation of function is also prion-variant dependent, indicating that only one of the two Sis1-prion functions may have been maintained in eukaryotic chaperone evolution., Author Summary Multiple neurodegenerative disorders such as Alzheimer's, Parkinson's and Creutzfeldt-Jakob disease are associated with the accumulation of fibrous protein aggregates collectively termed ‘amyloid.’ In the baker's yeast Saccharomyces cerevisiae, multiple proteins form intracellular amyloid aggregates known as yeast prions. Yeast prions minimally require a core set of chaperone proteins for stable propagation in yeast, including the J-protein Sis1, which appears to be required for the propagation of all yeast prions and functioning similarly in each case. Here we present evidence which challenges the notion of a universal function for Sis1 in prion propagation and asserts instead that Sis1's function in the maintenance of at least two prions, [RNQ +] and [PSI +], is distinct and mutually exclusive for some prion variants. We also find that the human homolog of Sis1, called Hdj1, has retained the ability to support some, but not all yeast prions, indicating a partial conservation of function. Because yeast chaperones have the ability to both bind and fragment amyloids in vivo, further investigations into these prion-specific properties of Sis1 and Hdj1 will likely lead to new insights into the biological management of protein misfolding.

Yeast prions are heritable amyloid aggregates of yeast proteins. Their propagation to subsequent cell generations is dependent upon their fragmentation by molecular chaperones for which mounting ev...

This year marks the 30th anniversary of the first description of the cellular distribution of actin within a yeast cell. Since then advances in both molecular genetics and imaging technologies have ensured research within these simple model organisms has blazed a trail in the field of actomyosin research. Many yeast proteins and their functions are functionally conserved in human cells. This, combined with experimental speed, minimal cost and ease of use make the yeasts extremely attractive model organisms for researching diverse cellular processes, including those involving actomyosin. In this chapter, current state-of-the-art fluorescence methodologies being applied to yeast actomyosin research, together with an honest appraisal of their limitations, such as the pitfalls that should be considered when fluorescently labelling proteins interacting within a dynamic cytoskeleton, will be discussed. Papers describing the established techniques developed for yeast localisation studies will be highlighted. This will provide the reader with an informed overview of the arsenal of imaging techniques available to the yeast actomyosin researcher and encourage them to consider novel ways these simple unicellular eukaryotes could be used to address their own research questions.

Identification of proteins from the mass spectra of peptide fragments generated by proteolytic cleavage using database searching has become one of the most powerful techniques in proteome science, capable of rapid and efficient protein identification. Using computer simulation, we have studied how the application of chemical derivatisation techniques may improve the efficiency of protein identification from mass spectrometric data. These approaches enhance ion yield and lead to the promotion of specific ions and fragments, yielding additional database search information. The impact of three alternative techniques has been assessed by searching representative proteome databases for both single proteins and simple protein mixtures. For example, by reliably promoting fragmentation of singly-charged peptide ions at aspartic acid residues after homoarginine derivatisation, 82% of yeast proteins can be unambiguously identified from a single typical peptide-mass datum, with a measured mass accuracy of 50 ppm, by using the associated secondary ion data. The extra search information also provides a means to confidently identify proteins in protein mixtures where only limited data are available. Furthermore, the inclusion of limited sequence information for the peptides can compensate and exceed the search efficiency available via high accuracy searches of around 5 ppm, suggesting that this is a potentially useful approach for simple protein mixtures routinely obtained from two-dimensional gels.

A protocol for the incorporation of SeMet into yeast proteins is described. Incorporation at a level of about 50% suffices for the location of Se sites in an anomalous difference Fourier map of the 0.5 MDa yeast RNA polymerase II. This shows the utility of the approach as an aid in the model-building of large protein complexes.

The two yeast proteins Rna14p and Rna15p form part of the cleavage/polyadenylation factor IA (CF IA) complex that is involved in the 3' processing of pre-mRNA. Association of the two proteins is mediated by a small C-terminal peptide from Rna14p and a region in Rna15p that corresponds to the hinge domain first identified within the human orthologue. Here I report the (1)H, (13)C and (15)N spectral assignments for a bacterially co-expressed heterodimer of Rna14p/Rna15p. Further analysis of secondary chemical shifts reveals that both peptides are predominantly α-helical within the complex.

Protein–protein interactions play pivotal roles in various aspects of the structural and functional organization of the cell, and their complete description is indispensable to thorough understanding of the cell. As an approach toward this goal, here we report a comprehensive system to examine two-hybrid interactions in all of the possible combinations between proteins of Saccharomyces cerevisiae . We cloned all of the yeast ORFs individually as a DNA-binding domain fusion (“bait”) in a MAT a strain and as an activation domain fusion (“prey”) in a MAT α strain, and subsequently divided them into pools, each containing 96 clones. These bait and prey clone pools were systematically mated with each other, and the transformants were subjected to strict selection for the activation of three reporter genes followed by sequence tagging. Our initial examination of ≈4 × 10 6 different combinations, constituting ≈10% of the total to be tested, has revealed 183 independent two-hybrid interactions, more than half of which are entirely novel. Notably, the obtained binary data allow us to extract more complex interaction networks, including the one that may explain a currently unsolved mechanism for the connection between distinct steps of vesicular transport. The approach described here thus will provide many leads for integration of various cellular functions and serve as a major driving force in the completion of the protein–protein interaction map.