New computational tools for the analysis of microbiome dynamics

EVENT : C3BI Seminars


Main speaker : Eran Halperin, from UCLA (Computer Science Department & Departments of Human Genetics, Biomathematics & Department of Anesthesiology) Date : 25-06-2019 at 11:00 am Location : Auditorium Jaques Monod – MONOD (66) ,Institut Pasteur, Paris


High-throughput microbiome analysis has become ubiquitous over the last few years. However, the interpretation of the data is often non-trivial and highly depends on the methodology used for the analysis. I will describe a few methods for the analysis of microbiome in contexts that are typical in such analyses. First, I will describe a new method for microbial source tracking, that is, finding the sources of a microbiome sample. I will demonstrate how using the method one can reach very different conclusions (that make more biological sense) than using previous methods. Specifically, I will show examples on the dynamics of gut microbiome in babies, and on the usage of source tracking as a tool for disease diagnosis. Second, I will discuss novel approaches for the analysis of time-series micorbiome data, and here too, I will show how this new approach results in new biological insights, particularly on the dependence of microbiome in the future on the current composition of the microbiome. The talk will be self contained, and I will not assume any expert knowledge in computer science or statistics.


Due to security policy in Institut Pasteur, please register before if you plan to come to this meeting

Interpreting the cancer genome through physical and functional models of the cancer cell

EVENT : C3BI Seminars


Main speaker : Trey Ideker, from UC San Diego – School of Medicine Date : 21-06-2019 at 02:00 pm Location : Jules Bordet room – METCHNIKOFF (67) ,Institut Pasteur, Paris


Dr. Ideker is a Professor of Medicine at UC San Diego. He is the Director of the National Resource for Network Biology, the San Diego Center for Systems Biology, and the Cancer Cell Map Initiative. He is a pioneer in using genome-scale measurements to construct network models of cellular processes and disease.

Recently we and other laboratories have launched the Cancer Cell Map Initiative (ccmi.org) and have been building momentum. The goal of the CCMI is to produce a complete map of the gene and protein wiring diagram of a cancer cell. We and others believe this map, currently missing, will be a critical component of any future system to decode a patient’s cancer genome. I will describe efforts along several lines: 1. Coalition building. We have made notable progress in building a coalition of institutions to generate the data, as well as to develop the computational methodology required to build and use the maps. 2. Development of technology for mapping gene-gene interactions rapidly using the CRISPR system. 3. Causal network maps connecting DNA mutations (somatic and germline, coding and noncoding) to the cancer events they induce downstream. 4. Development of software and database technology to visualize and store cancer cell maps. 5. A machine learning system for integrating the above data to create multi-scale models of cancer cells. In a recent paper by Ma et al., we have shown how a hierarchical map of cell structure can be embedded with a deep neural network, so that the model is able to accurately simulate the effect of mutations in genotype on the cellular phenotype.

Due to security policy in Institut Pasteur, please register before if you plan to come to this meeting

Evolution of information in HIV-1 protease

EVENT : C3BI Seminars


Main speaker : Chris Adami, from Michigan State University Date : 06-06-2019 at 02:00 pm Location : Auditorium Francois Jacob – BIME (26) ,Institut Pasteur, Paris


Highly-active anti-retroviral therapy has been extremely effective at maintaining low levels of viral load in HIV-infected individuals, but emerging drug resistance is threatening those gains. When therapy is interrupted even briefly, HIV can evolve resistance to one or multiple drugs. Understanding how to stop viral evolution is an important goal of current research. I use HIV-1 protease sequences from public databases to study the dynamics of evolution over a span of nearly ten years, to compare patterns of adaptation in populations that are drug-naive to those that have taken one or multiple protease inhibitors. Using information theory, I show that the amount of information stored in protease sequences of patients that are on drug therapy has been increasing over time, suggesting that they are adapting to the drugs. In comparison, there is no increase in information in the sequences of patients that are drug naive. However, for the virus the increase in information comes at a price: because most of the information is stored in correlations between residues, the sequences are evolving into a more rugged area of the fitness landscape, which could make further evolution more difficult. While the data up to 2006 do not suggest a slowing down of evolution, such a trend may exist in data from later years not analyzed here.


Due to security policy in Institut Pasteur, please register before if you plan to come to this meeting

A population genetic interpretation of complex trait architecture in humans

EVENT : C3BI Seminars


Main speaker : Guy Sella, from Department of Biological Sciences, Columbia University Date : 02-05-2019 at 02:00 pm Location : Auditorium Francois Jacob – BIME (26) ,Institut Pasteur, Paris


Human genome-wide association studies (GWASs) are revealing the genetic architecture of anthropomorphic and biomedical traits, i.e., the frequencies and effect sizes of variants that contribute to heritable variation in a trait. To interpret these findings, we need to understand how genetic architectures are shaped by basic population genetic processes—notably, by mutation, natural selection, and genetic drift. Because many complex traits are subject to stabilizing selection and genetic variation that affects one trait often affects many others, we model the genetic architecture of a focal trait that arises under stabilizing selection in a multidimensional trait space. We solve the model at steady state, to find that the distribution of variances contributed by loci identified in GWASs should be well approximated by a simple functional form that depends on a single parameter: the expected contribution to genetic variance of a strongly selected site affecting the trait. This prediction fits the findings of GWASs for height and body mass index (BMI) well, allowing us to make inferences about the degree of pleiotropy and mutational target size for these traits. Our findings help to explain why the GWAS for height explains more of the heritable variance than the similarly sized GWAS for BMI and to predict the increase in explained heritability with study size. Considering the demographic history of European populations, in which these GWASs were performed, we further find that most of the associations they identified likely involve mutations that arose during the Out-of-Africa bottleneck at sites with selection coefficients around s = 0.001.


Due to security policy in Institut Pasteur, please register before if you plan to come to this meeting

Using Systems Approaches to Understand the Mechanism of Disease

EVENT : C3BI Seminars


Main speaker : Nevan Krogan, from Quantitative Biosciences Institute , UC San Francisco, USA Date : 11-04-2019 at 02:00 pm Location : Auditorium Francois Jacob – BIME (26) ,Institut Pasteur, Paris


There is a wide gap between the generation of large-scale biological data sets and more-detailed, structural and mechanistic studies. However, recent work that explicitly combine data from systems and structural biological approaches is having a profound effect on our ability to predict how mutations and small molecules affect atomic-level mechanisms, disrupt systems-level networks and ultimately lead to changes in organismal fitness. Our group aims to create a stronger bridge between these areas primarily using three types of data: genetic interactions, protein-protein interactions and post-translational modifications.  Protein structural information helps to prioritize and functionally understand these large-scale datasets; conversely global, unbiasedly collected datasets helps inform the more mechanistic studies. Our efforts in this respect have been focused on three disease areas: cancer, infectious diseases and neuropsychiatric disorders. Our work has found remarkable similarities between these and other disease areas which are leading to novel therapeutic strategies.


Due to security policy in Institut Pasteur, please register before if you plan to come to this meeting

Human gut resistome

EVENT : C3BI Seminars


Main speaker : Amine Ghozlane, from HUB, C3BI Pasteur Date : 04-04-2019 at 02:00 pm Location : Auditorium Francois Jacob – BIME (26) ,Institut Pasteur, Paris


The intestinal microbiota is considered to be a major reservoir of antibiotic resistance determinants (ARDs) that could potentially be transferred to bacterial pathogens via mobile genetic elements. Yet, this assumption is poorly supported by empirical evidence due to the distant homologies between known ARDs (mostly from culturable bacteria) and ARDs from the intestinal microbiota. Consequently, an accurate census of intestinal ARDs (that is, the intestinal resistome) has not yet been fully determined. For this purpose, we developed and validated an annotation method (called pairwise comparative modelling) on the basis of a three-dimensional structure (homology comparative modelling), leading to the prediction of 6,095 ARDs in a catalogue of 3.9 million proteins from the human intestinal microbiota. We found that the majority of predicted ARDs (pdARDs) were distantly related to known ARDs (mean amino acid identity 29.8%) and found little evidence supporting their transfer between species. According to the composition of their resistome, we were able to cluster subjects from the MetaHIT cohort (n = 663) into six resistotypes that were connected to the previously described enterotypes. Finally, we found that the relative abundance of pdARDs was positively associated with gene richness, but not when subjects were exposed to antibiotics. Altogether, our results indicate that the majority of intestinal microbiota ARDs can be considered intrinsic to the dominant commensal microbiota and that these genes are rarely shared with bacterial pathogens.

Due to security policy in Institut Pasteur, please register before if you plan to come to this meeting

Bayesian matrix factorization for drug discovery and precision medicine

EVENT : C3BI Seminars


Main speaker : Yves Moreau, from Center for Computational Systems Biology, KU Leuven Date : 31-01-2019 at 02:00 pm Location : Auditorium Francois Jacob – BIME (26) ,Institut Pasteur, Paris


Matrix factorization/completion methods provide an attractive framework to handle sparsely observed data, also called “scarce” data. A typical setting for scarce data are is clinical diagnosis in a real-world setting. Not all possible symptoms (phenotype/biomarker/etc.) will have been checked for every patient. Deciding which symptom to check based on the already available information is at the heart of the diagnostic process. If genetic information about the patient is also available, it can serve as side information (covariates) to predict symptoms (phenotypes) for this patient. While a classification/regression setting is appropriate for this problem, it will typically ignore the dependencies between different tasks (i.e., symptoms). We have recently focused on a problem sharing many similarities with the diagnostic task: the prediction of biological activity of chemical compounds against drug targets, where only 0.1% to 1% of all compound-target pairs are measured. Matrix factorization searches for latent representations of compounds and targets that allow an optimal reconstruction of the observed measurements. These methods can be further combined with linear regression models to create multitask prediction models. In our case, fingerprints of chemical compounds are used as “side information” to predict target activity. By contrast with classical Quantitative Structure-Activity Relationship (QSAR) models, matrix factorization with side information naturally accommodates the multitask character of compound-target activity prediction. This methodology can be further extended to a fully Bayesian setting to handle uncertainty optimally, and our reformulation allows scaling up this MCMC scheme to millions of compounds, thousands of targets, and tens of millions of measurements, as demonstrated on a large industrial data set from a pharmaceutical company. We also show applications of this methodology to the prioritization of candidate disease genes and to the modeling of longitudinal patient trajectories. We have implemented our method as an open source Python/C++ library, called Macau, which can be applied to many modeling tasks, well beyond our original pharmaceutical setting. https://github.com/jaak-s/macau/tree/master/python/macau.


Due to security policy in Institut Pasteur, please register before if you plan to come to this meeting

Deciphering gene expression programs at single-cell resolution

EVENT : JOINT Seminar C3BI –  DPT DE Biologie du développement et cellules souches


Main speaker : Stein Aerts, from Laboratory of Computational Biology. KU Leuven Center for Human Genetics. VIB Center for Brain and Disease Research. Date : 15-02-2019 at 11:00 am Location : Jules Bordet room – METCHNIKOFF (67) ,Institut Pasteur, Paris


Single-cell technologies are revolutionising biology and provide new opportunities to trace genomic regulatory programs underlying cell fate. In this talk I will present several computational strategies for the analysis of single-cell RNA-seq and single-cell ATAC-seq data that exploit the genomic regulatory code, to guide the identification of transcription factors and cell states. I will illustrate these methods on several model systems, including the Drosophila brain. Finally I will discuss how single-cell analyses can contribute to cross-species comparisons of regulatory programs.

Prof. Stein Aerts has a multidisciplinary background in both bio-engineering and computer science. During his PhD he was trained in bioinformatics, and during his Postdoc he worked on the genomics of gene regulation in Drosophila. Stein now heads the Laboratory of Computational Biology at the VIB Center for Brain & Disease Research and the KU Leuven Department of Human Genetics. His lab focuses on deciphering the genomic regulatory code, using a combination of single-cell and machine-learning approaches. His most recent scientific contributions include new bioinformatics methods for the analysis of single-cell gene regulatory networks, namely SCENIC and cisTopic. Aerts co-founded the Fly Cell Atlas consortium and generated a single-cell atlas of the ageing Drosophila brain (scope.aertslab.org). Stein holds an ERC Consolidator Grant and was awarded the 2017 Prize for Bioinformatics and Computational Science from the Biotech Fund and the 2016 Astrazeneca Foundation Award Bioinformatics.

Due to security policy in Institut Pasteur, please register before if you plan to come to this meeting

A Polymer Physics View on Universal and Sequence-Specific Aspects of Chromosome Folding

EVENT : C3BI Seminars


Main speaker : Ralf Everaers, from Laboratoire Physique ENS Lyon (UMR CNRS 5672) Date : 17-01-2019 at 02:00 pm Location : Auditorium Francois Jacob – BIME (26) ,Institut Pasteur, Paris


Recent advances in genome-wide mapping and imaging techniques have strikingly improved the resolution at which nuclear genome folding can be analyzed and revealed numerous conserved features organizing the one-dimensional chromatin fiber into tridimensional nuclear domains. Understanding the underlying mechanisms and the link to gene regulation requires a crossdisciplinary approach that combines the new high-resolution techniques with computational modeling of chromatin and chromosomes. In the presentation I will discuss our current understanding of generic aspects of chromosome behavior during interphase. In collaboration with the Cavalli lab in Montpellier for the HiC experiments, we are using simulation techniques to explore their ability to explain the large scale chromosome folding in Drosophila nuclei during the course of development. We find that territory formation is fully described by the idea of topologically constrained relaxation of decondensing metaphase chromosomes. The characteristic signature of Rabl territories due to the memory of quasi-nematic chromosome alignment is visible during early stages of development, but disappears in late embryo nuclei. Compartimentalization of centromeric heterochromatin is well accounted for by co-polymer models with like-like attraction between hetero- and eu-chromatin. The additional distinction of a small number of epigenetic states allows to reasonably well predict the formation of (and interaction between) TADs.


Due to security policy in Institut Pasteur, please register before if you plan to come to this meeting

Hands-on microbiome data analysis: tools for understanding microbial communities in health and disease

EVENT : C3BI Training


Main speaker : Gregorio Iraola, from Institut Pasteur de Montevideo Date : 03-12-2018 at 09:00 am Location : Institut Pasteur de Montevideo


This course aims to provide the theoretical and practical concepts for standard bioinformatic analysis in the field of microbiome research. The course will focus on the application of state-of-the-art software tools for the analysis of environmental and host-associated microbiomes, with particular emphasis on understanding how they change or constitute a risk for human health. The course will have expert lectures and theoretical/practical data analysis sessions with real datasets.

 

STUDENT’S PRE-REQUISITES • Directed to post-graduation (M.Sc. or Ph.D.) students. • Basic concepts of high-throughput sequencing technologies. • Basic understanding of metagenomics and microbial ecology. • Basic skills in the Linux terminal.

 

TEACHERS

Institut Pasteur Montevideo

  • Chair: Gregorio Iraola
  • Pablo Fresia
  • Daniela Costa
  • Cecilia Salazar
  • Verónica Antelo
  • Ignacio Ferrés
  • Matias Giménez
Institut Pasteur Paris
  • Marie Lopez
  • Amine Ghozlane
  • Angèle Benard
    • INVITED SPEAKERS
      • Gianfranco Grompone, Discovery Microbiome, Nutrition & Health Science Lead, Lesaffre, France.
      • David Danko, Director of Bioinformatics, MetaSUB International Consortium, Weill Cornell Medicine, US
       

      DEADLINE APPLICATIONS October 19, 2018. Send your CV (one page) and letter of motivation to: antonio.borderia@pasteur.fr

      Flyer_Microbiome_health-course_Montevideo_2018