Nicolas Bertin

Evidence for dynamically organized modularity in the yeast protein-protein interaction network

In apparently scale-free protein–protein interaction networks, or ‘interactome’ networks, most proteins interact with few partners, whereas a small but significant proportion of proteins, the ‘hubs’, interact with many partners. Both biological and non-biological scale-free networks are particularly resistant to random node removal but are extremely sensitive to the targeted removal of hubs. A link between the potential scale-free topology of interactome networks and genetic robustness seems to exist, because knockouts of yeast genes encoding hubs are approximately threefold more likely to confer lethality than those of non-hubs. Here we investigate how hubs might contribute to robustness and other cellular properties for protein–protein interactions dynamically regulated both in time and in space. We uncovered two types of hub: ‘party’ hubs, which interact with most of their partners simultaneously, and ‘date’ hubs, which bind their different partners at different times or locations. Both in silico studies of network connectivity and genetic interactions described in vivo support a model of organized modularity in which date hubs organize the proteome, connecting biological processes—or modules —to each other, whereas party hubs function inside modules.

An atlas of active enhancers across human cell types and tissues

Enhancers control the correct temporal and cell-type-specific activation of gene expression in multicellular eukaryotes. Knowing their properties, regulatory activity and targets is crucial to understand the regulation of differentiation and homeostasis. Here we use the FANTOM5 panel of samples, covering the majority of human tissues and cell types, to produce an atlas of active, in vivo-transcribed enhancers. We show that enhancers share properties with CpG-poor messenger RNA promoters but produce bidirectional, exosome-sensitive, relatively short unspliced RNAs, the generation of which is strongly related to enhancer activity. The atlas is used to compare regulatory programs between different cells at unprecedented depth, to identify disease-associated regulatory single nucleotide polymorphisms, and to classify cell-type-specific and ubiquitous enhancers. We further explore the utility of enhancer redundancy, which explains gene expression strength rather than expression patterns. The online FANTOM5 enhancer atlas represents a unique resource for studies on cell-type-specific enhancers and gene regulation.

Predictive models of molecular machines involved in Caenorhabditis elegans early embryogenesis

Although numerous fundamental aspects of development have been uncovered through the study of individual genes and proteins, system-level models are still missing for most developmental processes. The first two cell divisions of Caenorhabditis elegans embryogenesis constitute an ideal test bed for a system-level approach. Early embryogenesis, including processes such as cell division and establishment of cellular polarity, is readily amenable to large-scale functional analysis. A first step toward a system-level understanding is to provide ‘first-draft’ models both of the molecular assemblies involved and of the functional connections between them. Here we show that such models can be derived from an integrated gene/protein network generated from three different types of functional relationship: protein interaction, expression profiling similarity and phenotypic profiling similarity, as estimated from detailed early embryonic RNA interference phenotypes systematically recorded for hundreds of early embryogenesis genes. The topology of the integrated network suggests that C. elegans early embryogenesis is achieved through coordination of a limited set of molecular machines. We assessed the overall predictive value of such molecular machine models by dynamic localization of ten previously uncharacterized proteins within the living embryo.

Gateways to the FANTOM5 promoter level mammalian expression atlas

The FANTOM5 project investigates transcription initiation activities in more than 1,000 human and mouse primary cells, cell lines and tissues using CAGE. Based on manual curation of sample information and development of an ontology for sample classification, we assemble the resulting data into a centralized data resource (http://fantom.gsc.riken.jp/5/). This resource contains web-based tools and data-access points for the research community to search and extract data related to samples, genes, promoter activities, transcription factors and enhancers across the FANTOM5 atlas.

Unexpected expression of α- and β-globin in mesencephalic dopaminergic neurons and glial cells

The mesencephalic dopaminergic (mDA) cell system is composed of two major groups of projecting cells in the substantia nigra (SN) (A9 neurons) and the ventral tegmental area (VTA) (A10 cells). A9 neurons form the nigrostriatal pathway and are involved in regulating voluntary movements and postural reflexes. Their selective degeneration leads to Parkinsons disease. Here, we report that gene expression analysis of A9 dopaminergic neurons (DA) identifies transcripts for α- and β-chains of hemoglobin (Hb). Globin immunoreactivity decorates the majority of A9 DA, a subpopulation of cortical and hippocampal astrocytes and mature oligodendrocytes. This pattern of expression was confirmed in different mouse strains and in rat and human. We show that Hb is expressed in the SN of human postmortem brain. By microarray analysis of dopaminergic cell lines overexpressing α- and β-globin chains, changes in genes involved in O2 homeostasis and oxidative phopshorylation were observed, linking Hb expression to mitochondrial function. Our data suggest that the most famed oxygen-carrying globin is not exclusively restricted to the blood, but it may play a role in the normal physiology of the brain and neurodegenerative diseases.

An atlas of human long non-coding RNAs with accurate 5′ ends

Long non-coding RNAs (lncRNAs) are largely heterogeneous and functionally uncharacterized. Here, using FANTOM5 cap analysis of gene expression (CAGE) data, we integrate multiple transcript collections to generate a comprehensive atlas of 27,919 human lncRNA genes with high-confidence 5′ ends and expression profiles across 1,829 samples from the major human primary cell types and tissues. Genomic and epigenomic classification of these lncRNAs reveals that most intergenic lncRNAs originate from enhancers rather than from promoters. Incorporating genetic and expression data, we show that lncRNAs overlapping trait-associated single nucleotide polymorphisms are specifically expressed in cell types relevant to the traits, implicating these lncRNAs in multiple diseases. We further demonstrate that lncRNAs overlapping expression quantitative trait loci (eQTL)-associated single nucleotide polymorphisms of messenger RNAs are co-expressed with the corresponding messenger RNAs, suggesting their potential roles in transcriptional regulation. Combining these findings with conservation data, we identify 19,175 potentially functional lncRNAs in the human genome.

Deep transcriptome profiling of mammalian stem cells supports a regulatory role for retrotransposons in pluripotency maintenance

The importance of microRNAs and long noncoding RNAs in the regulation of pluripotency has been documented; however, the noncoding components of stem cell gene networks remain largely unknown. Here we investigate the role of noncoding RNAs in the pluripotent state, with particular emphasis on nuclear and retrotransposon-derived transcripts. We have performed deep profiling of the nuclear and cytoplasmic transcriptomes of human and mouse stem cells, identifying a class of previously undetected stem cell–specific transcripts. We show that long terminal repeat (LTR)-derived transcripts contribute extensively to the complexity of the stem cell nuclear transcriptome. Some LTR-derived transcripts are associated with enhancer regions and are likely to be involved in the maintenance of pluripotency.

Systematic Interactome Mapping and Genetic Perturbation Analysis of a C. elegans TGF-β Signaling Network

To initiate a system-level analysis of C. elegans DAF-7/TGF-beta signaling, we combined interactome mapping with single and double genetic perturbations. Yeast two-hybrid (Y2H) screens starting with known DAF-7/TGF-beta pathway components defined a network of 71 interactions among 59 proteins. Coaffinity purification (co-AP) assays in mammalian cells confirmed the overall quality of this network. Systematic perturbations of the network using RNAi, both in wild-type and daf-7/TGF-beta pathway mutant animals, identified nine DAF-7/TGF-beta signaling modifiers, seven of which are conserved in humans. We show that one of these has functional homology to human SNO/SKI oncoproteins and that mutations at the corresponding genetic locus daf-5 confer defects in DAF-7/TGF-beta signaling. Our results reveal substantial molecular complexity in DAF-7/TGF-beta signal transduction. Integrating interactome maps with systematic genetic perturbations may be useful for developing a systems biology approach to this and other signaling modules.

Linking promoters to functional transcripts in small samples with nanoCAGE and CAGEscan

Large-scale sequencing projects have revealed an unexpected complexity in the origins, structures and functions of mammalian transcripts. Many loci are known to produce overlapping coding and noncoding RNAs with capped 5′ ends that vary in size. Methods to identify the 5′ ends of transcripts will facilitate the discovery of new promoters and 5′ ends derived from secondary capping events. Such methods often require high input amounts of RNA not obtainable from highly refined samples such as tissue microdissections and subcellular fractions. Therefore, we developed nano–cap analysis of gene expression (nanoCAGE), a method that captures the 5′ ends of transcripts from as little as 10 ng of total RNA, and CAGEscan, a mate-pair adaptation of nanoCAGE that captures the transcript 5′ ends linked to a downstream region. Both of these methods allow further annotation-agnostic studies of the complex human transcriptome.

Interactive visualization and analysis of large-scale sequencing datasets using ZENBU

217 clustering (Fig. 1c and Supplementary Figs. 3 and 4), and collation (Fig. 1h and Supplementary Fig. 3). With an understanding of these atomic operations, more advanced users can combine them into complex processing scripts. These customized scripts can also be saved and shared allowing efficient re-use of optimized analyses and views. To demonstrate some of ZENBU’s functionality, we show multiple views on the same underlying data, ENCODE RNA-seq2 experiments loaded in BAM format (Fig. 1). One of the more powerful views in this figure is the dynamic projection of RNA-seq reads onto GENCODE11 transcript models to calculate RPKM (reads per kilobase transcript per million reads) values (Fig. 1h). Transcripts are then colored by support and the RPKM expression table for these models can be downloaded in a variety of formats. A variation of the same script projects CAGE expression signal into the proximal promoter regions (±500 bp of 5ʹ end) of the same models (Supplementary Fig. 2). Furthermore, an implementation of the parametric clustering approach paraclu12 allows ZENBU to identify peaks of CAGE signal (Fig. 1c) and extract their corresponding expression values per experiment. Paraclu imposes minimal prior assumptions and is flexible enough for use with small RNA and ChIP-seq data by varying the clustering parameters (Supplementary Fig. 3 and 4). ZENBU also allows investigators to fine tune these settings and rapidly inspect the results on specific loci before genome-wide computation. A more detailed case study recapitulating an integrated ChIP-seq and RNA-seq analysis13 is included in Supplementary Note 1 to demonstrate the power of the complex scripting possible in ZENBU. To enable fast and reliable downloading of data genome-wide and to speed up visualization and processing, we implemented a track caching system, which To the Editor: The advance of sequencing technology has spurred an ever-growing body of sequence tag–based data from protocols such as RNA-seq, chromatin immunoprecipitation (ChIP)-seq, DNaseI-hypersensitive site sequencing (DHS)-seq and cap analysis of gene expression (CAGE). Large-scale consortia are applying standardized versions of these protocols across broad collections of samples1,2 to elucidate genomic function. Beyond this, the affordability of these systems and availability of sequencing services has made these technologies accessible to smaller laboratories focusing on individual biological systems. Data generation is only the beginning, however, and a substantial bottleneck for many labs is going from sequence data to biological insight, especially when the volume of data overwhelms standard paradigms for data visualization. Here, we present a web-based system, ZENBU, which addresses this problem by extending the functionality of the genome browser (Supplementary Fig. 1). For users with limited bioinformatics skills, ZENBU provides a suite of predefined views and data-processing scripts optimized for RNA-seq (Fig. 1), CAGE, short-RNA and ChIP-seq experiments (Supplementary Figs. 2–4). These are available simply upon uploading data as BAM (binary version of the sequence alignment/map)3 files. The system can also generate an optimized set of views for each of the above data types. ZENBU provides a rich selection of data manipulation capabilities, including quality filtering, signal thresholding, signal normalization, peak finding, annotation, collation of signal under peaks or transcript models and expression difference visualization across multiple experiments. We designed ZENBU with large-scale transcriptome projects in mind. In particular, the FANTOM5 project (unpublished data) required a system that would allow rapid incremental data loading, visualization, interpretation and downloading of thousands of deep CAGE, RNA-seq and small-RNA data sets as they were produced. We reviewed available systems4, including genome browsers, such as University of Tokyo Genome Browser (UTGB)5, Gbrowse6, University of Santa Cruz (UCSC) genome browser7, Ensembl8 and Integrative Genome Viewer (IGV)9, and data management tools, such as Biomart10, and found none with the full set of functions that we sought (see Supplementary Table 1 for comparison). Therefore, we developed ZENBU, a stable, fast, efficient and secure system that is flexible enough to allow customization of data filters, views and analyses. A key feature of ZENBU is that it allows the user to combine multiple experiments on demand (up to thousands of experiments) within any single track and interpret the data through linked views (Fig. 1). Upon combining multiple experiments, the genome browser view shows the combined genomic distribution of tags from these experiments within a given genomic interval (Fig. 1d), conversely the linked expression view shows the relative abundance of tags observed in this region across these experiments (shown as a histogram, Fig. 1j). The genome browser and expression view are ‘linked’, meaning that as the user interacts with one view the other is updated in real-time. This facilitates interactive exploration of the data because selecting features or regions within a browser track results in data for only that region being displayed in the expression view. And symmetrically, hiding specific experiments in the expression view updates the data displayed in the browser track. Data processing and complex views are achieved by a flexible on-demand scripting system based on data transformation and analysis modules. A selection of predefined combinations of simple operations are provided to perform generic tasks such as data normalization (Fig. 1b–d,h,i), data filtering (Supplementary Fig. 2), data Interactive visualization and analysis of large-scale sequencing datasets using ZENBU correspondence