Primary cilia are evolutionarily conserved cellular organelles that organize diverse signaling pathways1,2. Defects in the formation or function of primary cilia are associated with a spectrum of human diseases and developmental abnormalities3. Genetic screens in model organisms have discovered core machineries of cilium assembly and maintenance4. However, regulatory molecules that coordinate the biogenesis of primary cilia with other cellular processes, including cytoskeletal organization, vesicle trafficking and cell-cell adhesion, remain to be identified. Here we report the results of a functional genomic screen using RNA interference (RNAi) to identify human genes involved in ciliogenesis control. The screen identified 36 positive and 13 negative ciliogenesis modulators, which include molecules involved in actin dynamics and vesicle trafficking. Further investigation demonstrated that blocking actin assembly facilitates ciliogenesis by stabilizing the pericentrosomal preciliary compartment (PPC), a previously uncharacterized compact vesiculotubular structure storing transmembrane proteins destined for cilia during the early phase of ciliogenesis. PPC was labeled by recycling endosome markers. Moreover, knockdown of modulators that are involved in the endocytic recycling pathway affected the formation of PPC as well as ciliogenesis. Our results uncover a critical regulatory step that couples actin dynamics and endocytic recycling with ciliogenesis, and also provide potential target molecules for future study.

Summary The transcription factor ATF2 elicits oncogenic activities in melanoma and tumor suppressor activities in non-malignant skin cancer. Here we identify that ATF2 tumor suppressor function is determined by its ability to localize at the mitochondria, where it alters membrane permeability following genotoxic stress. The ability of ATF2 to reach the mitochondria is determined by PKCε, which directs ATF2 nuclear localization. Genotoxic stress attenuates PKCε effect on ATF2, enables ATF2 nuclear export and localization at the mitochondria, where it perturbs the HK1-VDAC1 complex, increases mitochondrial permeability and promotes apoptosis. Significantly, high levels of PKCε, as seen in melanoma cells, block ATF2 nuclear export and function at the mitochondria, thereby attenuating apoptosis following exposure to genotoxic stress. In melanoma tumor samples, high PKCε levels associates with poor prognosis. Overall, our findings provide the framework for understanding how subcellular localization enables ATF2 oncogenic or tumor suppressor functions.

The function of a protein is intimately tied to its subcellular localization. Although localizations have been measured for many yeast proteins through systematic GFP fusions, similar studies in other branches of life are still forthcoming. In the interim, various machine-learning methods have been proposed to predict localization using physical characteristics of a protein, such as amino acid content, hydrophobicity, side-chain mass and domain composition. However, there has been comparatively little work on predicting localization using protein networks. Here, we predict protein localizations by integrating an extensive set of protein physical characteristics over a protein's extended protein–protein interaction neighborhood, using a classification framework called ‘Divide and Conquer k-Nearest Neighbors’ (DC-kNN). These predictions achieve significantly higher accuracy than two well-known methods for predicting protein localization in yeast. Using new GFP imaging experiments, we show that the network-based approach can extend and revise previous annotations made from high-throughput studies. Finally, we show that our approach remains highly predictive in higher eukaryotes such as fly and human, in which most localizations are unknown and the protein network coverage is less substantial.

Supplementary data are available at Bioinformatics online.

The definition of protein–protein interactions (PPIs) in the natural cellular context is essential for properly understanding various biological processes. So far, however, most large-scale PPI analyses have not been performed in the natural cellular context. Here, we describe the construction of a Saccharomyces cerevisiae fusion library in which each endogenous gene is C-terminally tagged with the N-terminal fragment of Venus (VN) for a genome-wide bimolecular fluorescence complementation assay, a powerful technique for identifying PPIs in living cells. We illustrate the utility of the VN fusion library by systematically analyzing the interactome of the small ubiquitin-related modifier (SUMO) and provide previously unavailable information on the subcellular localization, types, and protease dependence of SUMO interactions. Our data set is highly complementary to the existing data sets and represents a useful resource for expanding the understanding of the physiological roles of SUMO. In addition, the VN fusion library provides a useful research tool that makes it feasible to systematically analyze PPIs in the natural cellular context.

Subcellular localization is one of the key functional characteristics of proteins. An automatic and efficient prediction method for the protein subcellular localization is highly required owing to the need for large-scale genome analysis. From a machine learning point of view, a dataset of protein localization has several characteristics: the dataset has too many classes (there are more than 10 localizations in a cell), it is a multi-label dataset (a protein may occur in several different subcellular locations), and it is too imbalanced (the number of proteins in each localization is remarkably different). Even though many previous works have been done for the prediction of protein subcellular localization, none of them tackles effectively these characteristics at the same time. Thus, a new computational method for protein localization is eventually needed for more reliable outcomes. To address the issue, we present a protein localization predictor based on D-SVDD (PLPD) for the prediction of protein localization, which can find the likelihood of a specific localization of a protein more easily and more correctly. Moreover, we introduce three measurements for the more precise evaluation of a protein localization predictor. As the results of various datasets which are made from the experiments of Huh et al. (2003), the proposed PLPD method represents a different approach that might play a complimentary role to the existing methods, such as Nearest Neighbor method and discriminate covariant method. Finally, after finding a good boundary for each localization using the 5184 classified proteins as training data, we predicted 138 proteins whose subcellular localizations could not be clearly observed by the experiments of Huh et al. (2003).

No abstract

The purpose of data description is to give a compact description of the target data that represents most of its characteristics. In a support vector data description (SVDD), the compact description of target data is given in a hyperspherical model, which is determined by a small portion of data called support vectors. Despite the usefulness of the conventional SVDD, however, it may not identify the optimal solution of target description especially when the support vectors do not have the overall characteristics of the target data. To address the issue in SVDD methodology, we propose a new SVDD by introducing new distance measurements based on the notion of a relative density degree for each data point in order to reflect the distribution of a given data set. Moreover, for a real application, we extend the proposed method for the protein localization prediction problem which is a multiclass and multilabel problem. Experiments with various real data sets show promising results.