Learning interpretable causal networks from very large datasets, application to 400,000 medical records of breast cancer patients.

Discovering causal effects is at the core of scientific investigation but remains challenging when only observational data are available. In practice, causal networks are difficult to learn and interpret, and limited to relatively small datasets. We report a more reliable and scalable causal discovery method (iMIIC), based on a general mutual information supremum principle, which greatly improves the precision of inferred causal relations while distinguishing genuine causes from putative and latent causal effects. We showcase iMIIC on synthetic and real-world healthcare data from 396,179 breast cancer patients from the US Surveillance, Epidemiology, and End Results program. More than 90% of predicted causal effects appear correct, while the remaining unexpected direct and indirect causal effects can be interpreted in terms of diagnostic procedures, therapeutic timing, patient preference or socio-economic disparity. iMIIC's unique capabilities open up new avenues to discover reliable and interpretable causal networks across a range of research fields.

A multistep computational approach reveals a neuro-mesenchymal cell population in the embryonic hematopoietic stem cell niche.

The first hematopoietic stem and progenitor cells (HSPCs) emerge in the Aorta-Gonad-Mesonephros (AGM) region of the mid-gestation mouse embryo. However, the precise nature of their supportive mesenchymal microenvironment remains largely unexplored. Here, we profiled transcriptomes of laser micro-dissected aortic tissues at three developmental stages and individual AGM cells. Computational analyses allowed the identification of several cell subpopulations within the E11.5 AGM mesenchyme, with the presence of a yet unidentified subpopulation characterized by the dual expression of genes implicated in adhesive or neuronal functions. We confirmed the identity of this cell subset as a neuro-mesenchymal population, through morphological and lineage tracing assays. Loss of function in the zebrafish confirmed that Decorin, a characteristic extracellular matrix component of the neuro-mesenchyme, is essential for HSPC development. We further demonstrated that this cell population is not merely derived from the neural crest, and hence, is a bona fide novel subpopulation of the AGM mesenchyme.

Interactive exploration of a global clinical network from a large breast cancer cohort.

Despite unprecedented amount of information now available in medical records, health data remain underexploited due to their heterogeneity and complexity. Simple charts and hypothesis-driven statistics can no longer apprehend the content of information-rich clinical data. There is, therefore, a clear need for powerful interactive visualization tools enabling medical practitioners to perceive the patterns and insights gained by state-of-the-art machine learning algorithms. Here, we report an interactive graphical interface for use as the front end of a machine learning causal inference server (MIIC), to facilitate the visualization and comprehension by clinicians of relationships between clinically relevant variables. The widespread use of such tools, facilitating the interactive exploration of datasets, is crucial both for data visualization and for the generation of research hypotheses. We demonstrate the utility of the MIIC interactive interface, by exploring the clinical network of a large cohort of breast cancer patients treated with neoadjuvant chemotherapy (NAC). This example highlights, in particular, the direct and indirect links between post-NAC clinical responses and patient survival. The MIIC interactive graphical interface has the potential to help clinicians identify actionable nodes and edges in clinical networks, thereby ultimately improving the patient care pathway.

Correlative AFM and fluorescence imaging demonstrate nanoscale membrane remodeling and ring-like and tubular structure formation by septins.

Septins are ubiquitous cytoskeletal filaments that interact with the inner plasma membrane and are essential for cell division in eukaryotes. In cellular contexts, septins are often localized at micrometric Gaussian curvatures, where they assemble onto ring-like structures. The behavior of budding yeast septins depends on their specific interaction with inositol phospholipids, enriched at the inner leaflet of the plasma membrane. Septin filaments are built from the non-polar self-assembly of short rods into filaments. However, the molecular mechanisms regulating the interplay with the inner plasma membrane and the resulting interaction with specific curvatures are not fully understood. In this report, we have imaged dynamical molecular assemblies of budding yeast septins on PIP2-containing supported lipid bilayers using a combination of high-speed AFM and correlative AFM-fluorescence microscopy. Our results clearly demonstrate that septins are able to bind to flat supported lipid bilayers and thereafter induce the remodeling of membranes. Short septin rods (octamers subunits) can indeed destabilize supported lipid bilayers and reshape the membrane to form 3D structures such as rings and tubes, demonstrating that long filaments are not necessary for septin-induced membrane buckling.

NeuriTES. Monitoring neurite changes through transfer entropy and semantic segmentation in bright-field time-lapse microscopy.

One of the most challenging frontiers in biological systems understanding is fluorescent label-free imaging. We present here the NeuriTES platform that revisits the standard paradigms of video analysis to detect unlabeled objects and adapt to the dynamic evolution of the phenomenon under observation. Object segmentation is reformulated using robust algorithms to assure regular cell detection and transfer entropy measures are used to study the inter-relationship among the parameters related to the evolving system. We applied the NeuriTES platform to the automatic analysis of neurites degeneration in presence of amyotrophic lateral sclerosis (ALS) and to the study of the effects of a chemotherapy drug on living prostate cancer cells (PC3) cultures. Control cells have been considered in both the two cases study. Accuracy values of 93% and of 92% are achieved, respectively. NeuriTES not only represents a tool for investigation in fluorescent label-free images but demonstrates to be adaptable to individual needs.

Inferring Gene Networks in Bone Marrow Hematopoietic Stem Cell-Supporting Stromal Niche Populations.

The cardinal property of bone marrow (BM) stromal cells is their capacity to contribute to hematopoietic stem cell (HSC) niches by providing mediators assisting HSC functions. In this study we first contrasted transcriptomes of stromal cells at different developmental stages and then included large number of HSC-supportive and non-supportive samples. Application of a combination of algorithms, comprising one identifying reliable paths and potential causative relationships in complex systems, revealed gene networks characteristic of the BM stromal HSC-supportive capacity and of defined niche populations of perivascular cells, osteoblasts, and mesenchymal stromal cells. Inclusion of single-cell transcriptomes enabled establishing for the perivascular cell subset a partially oriented graph of direct gene-to-gene interactions. As proof of concept we showed that R-spondin-2, expressed by the perivascular subset, synergized with Kit ligand to amplify ex vivo hematopoietic precursors. This study by identifying classifiers and hubs constitutes a resource to unravel candidate BM stromal mediators.

Learning clinical networks from medical records based on information estimates in mixed-type data.

The precise diagnostics of complex diseases require to integrate a large amount of information from heterogeneous clinical and biomedical data, whose direct and indirect interdependences are notoriously difficult to assess. To this end, we propose an efficient computational approach to simultaneously compute and assess the significance of multivariate information between any combination of mixed-type (continuous/categorical) variables. The method is then used to uncover direct, indirect and possibly causal relationships between mixed-type data from medical records, by extending a recent machine learning method to reconstruct graphical models beyond simple categorical datasets. The method is shown to outperform existing tools on benchmark mixed-type datasets, before being applied to analyze the medical records of eldery patients with cognitive disorders from La Pitié-Salpêtrière Hospital, Paris. The resulting clinical network visually captures the global interdependences in these medical records and some facets of clinical diagnosis practice, without specific hypothesis nor prior knowledge on any clinically relevant information. In particular, it provides some physiological insights linking the consequence of cerebrovascular accidents to the atrophy of important brain structures associated to cognitive impairment.

Mutations in calmodulin-binding domains of TRPV4/6 channels confer invasive properties to colon adenocarcinoma cells.

Transient receptor potential (TRP) channels form a family of polymodal cation channels gated by thermal, mechanical, or chemical stimuli, with many of them involved in the control of proliferation, apoptosis, or cell cycle. From an evolutionary point of view, TRP family is characterized by high conservation of duplicated genes originating from whole-genome duplication at the onset of vertebrates. The conservation of such "ohnolog" genes is theoretically linked to an increased probability of generating phenotypes deleterious for the organism upon gene mutation. We aimed to test experimentally the hypothesis that TRP mutations, in particular gain-of-function, could be involved in the generation of deleterious phenotypes involved in cancer, such as gain of invasiveness. Indeed, a number of TRP channels have been linked to cancer progression, and exhibit changes in expression levels in various types of cancers. However, TRP mutations in cancer have been poorly documented. We focused on 2 TRPV family members, TRPV4 and TRPV6, and studied the effect of putative gain-of-function mutations on invasiveness properties. TRPV channels have a C-terminal calmodulin-binding domain (CaMBD) that has important functions for regulating protein function, through different mechanisms depending on the channel (channel inactivation/potentiation, cytoskeleton regulation). We studied the effect of mutations mimicking constitutive phosphorylation in TRPV4 and TRPV6 CaMBDs: TRPV4 S823D, S824D and T813D, TRPV6 S691D, S692D and T702. We found that most of these mutants induced a strong gain of invasiveness of colon adenocarcinoma SW480 cells, both for TRPV4 and TRPV6. While increased invasion with TRPV6 S692D and T702D mutants was correlated to increased mutant channel activity, it was not the case for TRPV4 mutants, suggesting different mechanisms with the same global effect of gain in deleterious phenotype. This highlights the potential importance to search for TRP mutations involved in cancer.

OHNOLOGS v2: a comprehensive resource for the genes retained from whole genome duplication in vertebrates.

All vertebrates including human have evolved from an ancestor that underwent two rounds of whole genome duplication (2R-WGD). In addition, teleost fish underwent an additional third round of genome duplication (3R-WGD). The genes retained from these genome duplications, so-called ohnologs, have been instrumental in the evolution of vertebrate complexity, development and susceptibility to genetic diseases. However, the identification of vertebrate ohnologs has been challenging, due to lineage specific genome rearrangements since 2R- and 3R-WGD. We previously identified vertebrate ohnologs using a novel synteny comparison across multiple genomes. Here, we refine and apply this approach on 27 vertebrate genomes to identify ohnologs from both 2R- and 3R-WGD, while taking into account the phylogenetically biased sampling of available species. We assemble vertebrate ohnolog pairs and families in an expanded OHNOLOGS v2 database. We find that teleost fish have retained more 2R-WGD ohnologs than mammals and sauropsids, and that these 2R-ohnologs have retained significantly more ohnologs from the subsequent 3R-WGD than genes without 2R-ohnologs. Interestingly, species with fewer extant genes, such as sauropsids, have retained similar or higher proportions of ohnologs. OHNOLOGS v2 should allow deeper evolutionary genomic analysis of the impact of WGD on vertebrates and can be freely accessed at http://ohnologs.curie.fr.

Membrane reshaping by micrometric curvature sensitive septin filaments.

Septins are cytoskeletal filaments that assemble at the inner face of the plasma membrane. They are localized at constriction sites and impact membrane remodeling. We report in vitro tools to examine how yeast septins behave on curved and deformable membranes. Septins reshape the membranes of Giant Unilamellar Vesicles with the formation of periodic spikes, while flattening smaller vesicles. We show that membrane deformations are associated to preferential arrangement of septin filaments on specific curvatures. When binding to bilayers supported on custom-designed periodic wavy patterns displaying positive and negative micrometric radii of curvatures, septin filaments remain straight and perpendicular to the curvature of the convex parts, while bending negatively to follow concave geometries. Based on these results, we propose a theoretical model that describes the deformations and micrometric curvature sensitivity observed in vitro. The model captures the reorganizations of septin filaments throughout cytokinesis in vivo, providing mechanistic insights into cell division.

MIIC online: a web server to reconstruct causal or non-causal networks from non-perturbative data.

Summary: We present a web server running the MIIC algorithm, a network learning method combining constraint-based and information-theoretic frameworks to reconstruct causal, non-causal or mixed networks from non-perturbative data, without the need for an a priori choice on the class of reconstructed network. Starting from a fully connected network, the algorithm first removes dispensable edges by iteratively subtracting the most significant information contributions from indirect paths between each pair of variables. The remaining edges are then filtered based on their confidence assessment or oriented based on the signature of causality in observational data. MIIC online server can be used for a broad range of biological data, including possible unobserved (latent) variables, from single-cell gene expression data to protein sequence evolution and outperforms or matches state-of-the-art methods for either causal or non-causal network reconstruction. Availability and implementation: MIIC online can be freely accessed at https://miic.curie.fr. Supplementary information: Supplementary data are available at Bioinformatics online.

Learning causal networks with latent variables from multivariate information in genomic data.

Learning causal networks from large-scale genomic data remains challenging in absence of time series or controlled perturbation experiments. We report an information- theoretic method which learns a large class of causal or non-causal graphical models from purely observational data, while including the effects of unobserved latent variables, commonly found in many genomic datasets. Starting from a complete graph, the method iteratively removes dispensable edges, by uncovering significant information contributions from indirect paths, and assesses edge-specific confidences from randomization of available data. The remaining edges are then oriented based on the signature of causality in observational data. The approach and associated algorithm, miic, outperform earlier methods on a broad range of benchmark networks. Causal network reconstructions are presented at different biological size and time scales, from gene regulation in single cells to whole genome duplication in tumor development as well as long term evolution of vertebrates. Miic is publicly available at https://github.com/miicTeam/MIIC.

3off2: A network reconstruction algorithm based on 2-point and 3-point information statistics.

BACKGROUND: The reconstruction of reliable graphical models from observational data is important in bioinformatics and other computational fields applying network reconstruction methods to large, yet finite datasets. The main network reconstruction approaches are either based on Bayesian scores, which enable the ranking of alternative Bayesian networks, or rely on the identification of structural independencies, which correspond to missing edges in the underlying network. Bayesian inference methods typically require heuristic search strategies, such as hill-climbing algorithms, to sample the super-exponential space of possible networks. By contrast, constraint-based methods, such as the PC and IC algorithms, are expected to run in polynomial time on sparse underlying graphs, provided that a correct list of conditional independencies is available. Yet, in practice, conditional independencies need to be ascertained from the available observational data, based on adjustable statistical significance levels, and are not robust to sampling noise from finite datasets. RESULTS: We propose a more robust approach to reconstruct graphical models from finite datasets. It combines constraint-based and Bayesian approaches to infer structural independencies based on the ranking of their most likely contributing nodes. In a nutshell, this local optimization scheme and corresponding 3off2 algorithm iteratively "take off" the most likely conditional 3-point information from the 2-point (mutual) information between each pair of nodes. Conditional independencies are thus derived by progressively collecting the most significant indirect contributions to all pairwise mutual information. The resulting network skeleton is then partially directed by orienting and propagating edge directions, based on the sign and magnitude of the conditional 3-point information of unshielded triples. The approach is shown to outperform both constraint-based and Bayesian inference methods on a range of benchmark networks. The 3off2 approach is then applied to the reconstruction of the hematopoiesis regulation network based on recent single cell expression data and is found to retrieve more experimentally ascertained regulations between transcription factors than with other available methods. CONCLUSIONS: The novel information-theoretic approach and corresponding 3off2 algorithm combine constraint-based and Bayesian inference methods to reliably reconstruct graphical models, despite inherent sampling noise in finite datasets. In particular, experimentally verified interactions as well as novel predicted regulations are established on the hematopoiesis regulatory networks based on single cell expression data.

Identification of Ohnolog Genes Originating from Whole Genome Duplication in Early Vertebrates, Based on Synteny Comparison across Multiple Genomes.

Whole genome duplications (WGD) have now been firmly established in all major eukaryotic kingdoms. In particular, all vertebrates descend from two rounds of WGDs, that occurred in their jawless ancestor some 500 MY ago. Paralogs retained from WGD, also coined 'ohnologs' after Susumu Ohno, have been shown to be typically associated with development, signaling and gene regulation. Ohnologs, which amount to about 20 to 35% of genes in the human genome, have also been shown to be prone to dominant deleterious mutations and frequently implicated in cancer and genetic diseases. Hence, identifying ohnologs is central to better understand the evolution of vertebrates and their susceptibility to genetic diseases. Early computational analyses to identify vertebrate ohnologs relied on content-based synteny comparisons between the human genome and a single invertebrate outgroup genome or within the human genome itself. These approaches are thus limited by lineage specific rearrangements in individual genomes. We report, in this study, the identification of vertebrate ohnologs based on the quantitative assessment and integration of synteny conservation between six amniote vertebrates and six invertebrate outgroups. Such a synteny comparison across multiple genomes is shown to enhance the statistical power of ohnolog identification in vertebrates compared to earlier approaches, by overcoming lineage specific genome rearrangements. Ohnolog gene families can be browsed and downloaded for three statistical confidence levels or recompiled for specific, user-defined, significance criteria at http://ohnologs.curie.fr/. In the light of the importance of WGD on the genetic makeup of vertebrates, our analysis provides a useful resource for researchers interested in gaining further insights on vertebrate evolution and genetic diseases.

On the retention of gene duplicates prone to dominant deleterious mutations.

Recent studies have shown that gene families from different functional categories have been preferentially expanded either by small scale duplication (SSD) or by whole-genome duplication (WGD). In particular, gene families prone to dominant deleterious mutations and implicated in cancers and other genetic diseases in human have been greatly expanded through two rounds of WGD dating back from early vertebrates. Here, we strengthen this intriguing observation, showing that human oncogenes involved in different primary tumors have retained many WGD duplicates compared to other human genes. In order to rationalize this evolutionary outcome, we propose a consistent population genetics model to analyze the retention of SSD and WGD duplicates taking into account their propensity to acquire dominant deleterious mutations. We solve a deterministic haploid model including initial duplicated loci, their retention through sub-functionalization or their neutral loss-of-function or deleterious gain-of-function at one locus. Extensions to diploid genotypes are presented and population size effects are analyzed using stochastic simulations. The only difference between the SSD and WGD scenarios is the initial number of individuals with duplicated loci. While SSD duplicates need to spread through the entire population from a single individual to reach fixation, WGD duplicates are de facto fixed in the small initial post-WGD population arising through the ploidy incompatibility between post-WGD individuals and the rest of the pre-WGD population. WGD duplicates prone to dominant deleterious mutations are then shown to be indirectly selected through purifying selection in post-WGD species, whereas SSD duplicates typically require positive selection. These results highlight the long-term evolution mechanisms behind the surprising accumulation of WGD duplicates prone to dominant deleterious mutations and are shown to be consistent with cancer genome data on the prevalence of human oncogenes with WGD duplicates.

On the expansion of "dangerous" gene repertoires by whole-genome duplications in early vertebrates.

The emergence and evolutionary expansion of gene families implicated in cancers and other severe genetic diseases is an evolutionary oddity from a natural selection perspective. Here, we show that gene families prone to deleterious mutations in the human genome have been preferentially expanded by the retention of "ohnolog" genes from two rounds of whole-genome duplication (WGD) dating back from the onset of jawed vertebrates. We further demonstrate that the retention of many ohnologs suspected to be dosage balanced is in fact indirectly mediated by their susceptibility to deleterious mutations. This enhanced retention of "dangerous" ohnologs, defined as prone to autosomal-dominant deleterious mutations, is shown to be a consequence of WGD-induced speciation and the ensuing purifying selection in post-WGD species. These findings highlight the importance of WGD-induced nonadaptive selection for the emergence of vertebrate complexity, while rationalizing, from an evolutionary perspective, the expansion of gene families frequently implicated in genetic disorders and cancers.

A boost for the emerging field of RNA nanotechnology.

This Nano Focus article highlights recent advances in RNA nanotechnology as presented at the First International Conference of RNA Nanotechnology and Therapeutics, which took place in Cleveland, OH, USA (October 23-25, 2010) ( http://www.eng.uc.edu/nanomedicine/RNA2010/ ), chaired by Peixuan Guo and co-chaired by David Rueda and Scott Tenenbaum. The conference was the first of its kind to bring together more than 30 invited speakers in the frontier of RNA nanotechnology from France, Sweden, South Korea, China, and throughout the United States to discuss RNA nanotechnology and its applications. It provided a platform for researchers from academia, government, and the pharmaceutical industry to share existing knowledge, vision, technology, and challenges in the field and promoted collaborations among researchers interested in advancing this emerging scientific discipline. The meeting covered a range of topics, including biophysical and single-molecule approaches for characterization of RNA nanostructures; structure studies on RNA nanoparticles by chemical or biochemical approaches, computation, prediction, and modeling of RNA nanoparticle structures; methods for the assembly of RNA nanoparticles; chemistry for RNA synthesis, conjugation, and labeling; and application of RNA nanoparticles in therapeutics. A special invited talk on the well-established principles of DNA nanotechnology was arranged to provide models for RNA nanotechnology. An Administrator from National Institutes of Health (NIH) National Cancer Institute (NCI) Alliance for Nanotechnology in Cancer discussed the current nanocancer research directions and future funding opportunities at NCI. As indicated by the feedback received from the invited speakers and the meeting participants, this meeting was extremely successful, exciting, and informative, covering many groundbreaking findings, pioneering ideas, and novel discoveries.

A nanostructure made of a bacterial noncoding RNA.

Natural RNAs, unlike many proteins, have never been reported to form extended nanostructures, despite their wide variety of cellular functions. This is all the more striking, as synthetic DNA and RNA forming large nanostructures have long been successfully designed. Here, we show that DsrA, a 87-nt noncoding RNA of Escherichia coli, self-assembles into a hierarchy of nanostructures through antisense interactions of three contiguous self-complementary regions. Yet, the extended nanostructures, observed using atomic force microscopy (AFM) and fluorescence microscopy, are easily disrupted into >100 nm long helical bundles of DsrA filaments, including hundreds of DsrA monomers, and are surprisingly resistant to heat and urea denaturation. Molecular modeling demonstrates that this structural switch of DsrA nanostructures into filament bundles results from the relaxation of stored torsional constraints and suggests possible implications for DsrA regulatory function.