- statslausanne - lists.sib.swiss

Statistical forum on Thursday March 29 at 11:00
by Rousson Valentin 12 Mar '18

12 Mar '18

Dear all, The next seminar in our series of statistics talks for people interested in quantitative methods in biology and medicine will take place on Thursday 29 March 2018 at 11:00, in the Salle Delachaux (Biopôle 2, first floor, Route de la Corniche 10, 1010 Lausanne, M2: Vennes). The speaker will be the Dr. Jack Bowden from the "Bristol Population Health Science Institute" and the "Bristol Medical School" of the University of Bristol, UK. Title: Recent advances in two-sample summary data Mendelian randomization Abstract: The explosion in publicly available data from genome-wide association studies is accelerating the use of Mendelian randomization in epidemiology. Summary data estimates of genetic association from large numbers of variants are now being synthesized to yield causal estimates within the framework of a meta-analysis. In a short time, the field has seen a dramatic rise in the power for testing causal hypotheses. There is a justified concern that when large numbers of genetic variants are included in a Mendelian randomization analyses, with many lacking a firm biological basis for their association with the exposure, a sizeable proportion of these variants are likely to be invalid instrumental variables. Specifically, there is a real danger that many variants will exert pleiotropic effects on the outcome not through the exposure of interest, thus biasing the analysis and leading to incorrect inferences. In this talk I will start by giving a whistle-stop tour of methodological approaches to address the issue of pleiotropy in Mendelian randomization. I will finish by discussing very recent work on the simultaneous adjustment for weak and pleiotropic instruments within MR. For further reading see: https://www.biorxiv.org/content/early/2018/02/27/159442 https://arxiv.org/abs/1801.09652 https://www.biorxiv.org/content/early/2017/10/11/200378 Hope to see you there! Valentin (for the organizers: Zoltán Kutalik, Valentin Rousson, Frédéric Schütz) CHUV centre hospitalier universitaire vaudois Valentin ROUSSON - Professeur associé, mathematicien/statisticien Département universitaire de medecine et santé communautaires (DUMSC) Médecine sociale et préventive (IUMSP) Biostatistique et méthodes quantitatives BIO 2/2ème/103 Rte de la Corniche 10, CH - 1010 Lausanne +41 (0)21 314 73 28 TEL Valentin.Rousson(a)chuv.ch<mailto:Valentin.Rousson@chuv.ch> www.chuv.ch<http://www.chuv.ch/>

1 0

Statistical forum on Thursday January 11 at 14:00
by Rousson Valentin 04 Jan '18

04 Jan '18

Dear all, The next seminar in our series of statistics talks for people interested in quantitative methods in biology and medicine will take place on Thursday 11 January 2018 at 14:00, in the conference room C04 situated on the floor -1 of the Biopôle 1 building (also known as SC-B building, see the enclosed map) located on the route de la Corniche 10, 1066 Epalinges (M2 stop: Les Croisettes). The speakers will be Prof. Marcel Zwahlen and Dr. Adrian Spoerri from the Institute for Social and Preventive Medicine of the University of Berne. Two articles about this presentation are enclosed to this email. Title: Record linkage: deterministic versus probabilistic versus data protection issues Abstract: Record linkage of existing individual health care data is an efficient way to answer important epidemiological research questions. Reuse of individual health-related data faces several problems: Either a unique personal identifier, like social security number, is not available or non-unique person identifiable information, like names, are privacy protected and cannot be accessed. A solution to protect privacy in probabilistic record linkages is to encrypt these sensitive information. Unfortunately, encrypted hash codes of two names differ completely if the plain names differ only by a single character. Therefore, standard encryption methods cannot be applied. To overcome these challenges, and comply certain encryption methods exist which allow to preserve data protection (Privacy Preserving Probabilistic Record Linkage (P3RL) methods). We will present the principles of probabilistic record linkage and of the P3RL methods. Happy new year and hope to see you there! Valentin (for the organizers: Zoltán Kutalik, Valentin Rousson, Frédéric Schütz)

1 0

This afternoon at 15:30: Presentation by Stephen Senn on Precision Medicine
by Frédéric Schütz 22 Nov '17

22 Nov '17

Dear all, quick reminder that this afternoon, at 15:30, the SIB and the CIG will host statistician Stephen Senn (Head of Competence Center for Methodology and Statistics (CCMS), Luxembourg Institute of Health) who will talk about: Is Precision Medicine Terminally Ill? The presentation will take place in Auditoire A of the Génopode Building, on the Dorigny campus of the University of Lausanne. The presentation will be broadcast live at https://collab.switch.ch/sib-cbss/ It will also be available later on the SIB Youtube channel at https://www.youtube.com/channel/UCPo4ED_WAKjwQ878cca6_oQ Hope to see you there ! Frédéric Abstract: For at least twenty years we have been promised that a revolution in medical care, based mainly on genomics, but also on various other 'omics', is imminent. Precise diagnosis will be matched by tailor-made medicine and put an end to mass solutions doled out to crudely assembled disease groups. In those 20 years, the relevant vocabulary has been reinvented several times moving from personalized, to precision to high-definition medicine. However, despite some notable successes the theranostic revolution has not taken place and the reality continues to fall far short of the promise. I shall argue that one reason is that clinical trials are regularly misunderstood and all too frequently badly analysed. In particular, an obsession with numbers needed to treat based on arbitrary dichotomies has exaggerated the extent to which patients differ in response and hence the scope for personalising therapies. I shall suggest that smarter design and analysis of trial is needed and that it is quite probable that the scope for personalised medicine is less than commonly supposed. The main task for drug development will remain finding treatments that are good on average. ---

1 0

Statistical forum on Wednesday 22 November 15:30
by Frédéric Schütz 08 Nov '17

08 Nov '17

Dear all, The next seminar in our series of statistics talks for people interested in quantitative methods in biology and medicine will take place on Wednesday 22 November 2017, 15:30, in Auditoire A of the Génopode Building (Dorigny campus). The speaker will be Stephen Senn (Head of Competence Center for Methodology and Statistics (CCMS), Luxembourg Institute of Health) Title: Is Precision Medicine Terminally Ill? For at least twenty years we have been promised that a revolution in medical care, based mainly on genomics, but also on various other 'omics', is imminent. Precise diagnosis will be matched by tailor-made medicine and put an end to mass solutions doled out to crudely assembled disease groups. In those 20 years, the relevant vocabulary has been reinvented several times moving from personalized, to precision to high-definition medicine. However, despite some notable successes the theranostic revolution has not taken place and the reality continues to fall far short of the promise. I shall argue that one reason is that clinical trials are regularly misunderstood and all too frequently badly analysed. In particular, an obsession with numbers needed to treat based on arbitrary dichotomies has exaggerated the extent to which patients differ in response and hence the scope for personalising therapies. I shall suggest that smarter design and analysis of trial is needed and that it is quite probable that the scope for personalised medicine is less than commonly supposed. The main task for drug development will remain finding treatments that are good on average. --- Hope to see you there! Frédéric

1 0

Statistical forum on Friday October 13 at 11:00
by Rousson Valentin 02 Nov '17

02 Nov '17

Dear all, The next seminar in our series of statistics talks for people interested in quantitative methods in biology and medicine will take place on Friday 13 October 2017 at 11:00, in the Salle Delachaux (Biopôle 2, first floor, Route de la Corniche 10, 1010 Lausanne, M2: Vennes). The speaker will be Prof. Eva Cantoni (Research Center for Statistics and Geneva School of Economics and Management, University of Geneva). Titre: A random-effects hurdle model for predicting bycatch of endangered marine species Abstract: Understanding and reducing the incidence of accidental bycatch, particularly for vulnerable species such as sharks, is a major challenge for contemporary fisheries management worldwide. Bycatch data, most often collected by at-sea observers during fishing trips, are clustered by trip and/or vessel and typically involve a large number of zero counts and very few positive counts. Though hurdle models are very popular for count data with excess zeros, models for clustered forms have received far less attention. Here we present a novel random-effects hurdle model for bycatch data that makes available accurate estimates of bycatch probabilities as well as other cluster-specific targets. These are essential for informing conservation and management decisions as well as for identifying bycatch hotspots, often considered the first step in attempting to protect endangered marine species. We validate our methodology through simulation and use it to analyse bycatch data on critically endangered hammerhead sharks from the U.S. National Marine Fisheries Service Pelagic Observer Program. Hope to see you there! Valentin (for the organizers: Zoltán Kutalik, Valentin Rousson, Frédéric Schütz) CHUV centre hospitalier universitaire vaudois Valentin ROUSSON - Professeur associé, mathematicien/statisticien Département universitaire de medecine et santé communautaires (DUMSC) Médecine sociale et préventive (IUMSP) Biostatistique et méthodes quantitatives BIO 2/2ème/103 Rte de la Corniche 10, CH - 1010 Lausanne +41 (0)21 314 73 28 TEL Valentin.Rousson(a)chuv.ch<mailto:Valentin.Rousson@chuv.ch> www.chuv.ch<http://www.chuv.ch/>

1 1

Statistical forum on August 24 at 11:00
by Rousson Valentin 08 Aug '17

08 Aug '17

Dear all, The next seminar in our series of statistics talks for people interested in quantitative methods in biology and medicine will take place on Thursday 24 August 2017 at 11:00. Exceptionally, it will take place in the room C4 at the level -1 from the Biopôle 1 (Route de la Corniche 2, 1066 Epalinges, M2 stop: Les Croisettes). The speaker will be Prof. John Braun (University of British Columbia, Canada). Title: A Bootstrapped Version of the Prometheus Wildland Fire Growth Model Abstract: The Prometheus Fire Growth Model is a deterministic wildfire simulator used to predict the growth of a wildifre, in Canada and other countries. Given weather, topographical and fuel information, the simulated fire front is plotted at equally spaced times. Unpredictability of fire behaviour makes deterministic predictions inaccurate. This talk will describe a risk analysis study undertaken using Prometheus applied to random weather streams, pointing out limitations with that approach and motivating an alternative viewpoint, based on bootstrapping. By statistically modelling the data to which the Prometheus model equations are fit, it is possible to obtain a distribution of fire front predictions. This approach allows us to estimate the probability that a growing fire will eventually burn a particular location. Repeated stochastic simulation is not required, so probability contours require no more computing time than deterministic contours. We conclude with a discussion of the implications for improved fire risk analysis. Hope to see you there! Valentin (for the organizers: Zoltán Kutalik, Valentin Rousson, Frédéric Schütz) CHUV centre hospitalier universitaire vaudois Valentin ROUSSON - Professeur associé, mathematicien/statisticien Département universitaire de medecine et santé communautaires (DUMSC) Médecine sociale et préventive (IUMSP) Biostatistique et méthodes quantitatives BIO 2/2ème/103 Rte de la Corniche 10, CH - 1010 Lausanne +41 (0)21 314 73 28 TEL Valentin.Rousson(a)chuv.ch<mailto:Valentin.Rousson@chuv.ch> www.chuv.ch<http://www.chuv.ch/>

1 0

Statistical forum on Thursday June 8 at 11:00
by Rousson Valentin 24 May '17

24 May '17

Dear all, The next seminar in our series of statistics talks for people interested in quantitative methods in biology and medicine will take place on this Thursday 8 June 2017 at 11:00, in the Salle Delachaux (Biopôle 2, first floor, Route de la Corniche 10, 1010 Lausanne, M2: Vennes). The speaker will be Dr. Aaron McDaid (group of Zoltán Kutalik, IUMSP, University Hospital Lausanne). Title: Bayesian association scan reveals loci associated with human lifespan and linked biomarkers Abstract: The enormous variations in human lifespan are in part due to a myriad of sequence variants, only a few of which have been revealed to date. Since many life-shortening events are related to diseases, we developed a Mendelian randomization-based method combining 58 disease-related GWA studies to derive longevity priors for all HapMap SNPs. A Bayesian association scan, informed by these priors, for parental age of death in the UK Biobank study (n=116,279) revealed 16 independent SNPs with significant Bayes factor at a 5% false discovery rate (FDR). Eleven of them replicate (5% FDR) in five independent longevity studies combined; all but three are depleted of the life-shortening alleles in older Biobank participants. Further analysis revealed that brain expression levels of nearby genes (RBM6, SULT1A1, CHRNA5) might be causally implicated in longevity. Gene-expression and caloric restriction experiments in model organisms confirm the conserved role for RBM6 and SULT1A1 in modulating lifespan. Hope to see you there! Valentin (for the organizers: Zoltán Kutalik, Valentin Rousson, Frédéric Schütz)

1 0

Statistical forum on March 30 at 11:00
by Rousson Valentin 01 May '17

01 May '17

Dear all, The next seminar in our series of statistics talks for people interested in quantitative methods in biology and medicine will take place on Thursday 30 March 2017 at 11:00, in the Salle Delachaux (Biopôle 2, first floor, Route de la Corniche 10, 1010 Lausanne, M2: Vennes). The speaker will be Serguei Rouzinov (Faculty of Social and Political Sciences, University of Lausanne) Title: Beyond classical tests: A distributional approach to missingness. Abstract: Missing data occur in almost all surveys, and to handle them correctly, it is essential that we know their type. Missing data are generally divided into three types (or generating mechanisms): missing completely at random, missing at random, and missing not at random. In addition, the generating mechanism can be a mixture of these single mechanisms. The first step to understanding their type generally consists of testing whether or not the missing data are missing completely at random. Several tests have been developed for that purpose, but they are not appropriate for mixed missing data mechanisms and have difficulties when dealing with non-continuous variables. Our approach tests whether the missing data are missing completely at random or missing at random with a regression model and a distribution test. Formally, for a variable with missing data, we compare the predictions of the regression model given for observed data with those given for unobserved data. Our simulation results show that compared to the standard Little test, our method is at least as powerful for single mechanisms but clearly better for mixed mechanisms. This is a joint work with Prof. André Berchtold Happy new year and hope to see you there! Valentin (for the organizers: Jérôme Goudet, Valentin Rousson, Frédéric Schütz)

1 1

statistical forum on January 19 at 15:00
by Rousson Valentin 09 Jan '17

09 Jan '17

Dear all, The next seminar in our series of statistics talks for people interested in quantitative methods in biology and medicine will take place on Thursday 19 January 2017 at 15:00 (unusual time!), in the Salle Delachaux (Biopôle 2, first floor, Route de la Corniche 10, 1010 Lausanne, M2: Vennes). The speaker will be Dr. Konstantin Popadin (Centre for Integrative Genomics, University of Lausanne) Title: Main targets of the negative selection in the human genome Abstract: All organisms harbor many slightly-deleterious (SD) variants in their genomes. For example, every human genome has approximately one thousand point SD coding variants, several thousand point SD regulatory variants, and dozens of SD copy number variants. Even though most of these mutations individually have small effects, their cumulative effect (mutation load) is expected to affect health status and reduce the fitness of an individual. Thus the mutational load is assumed to be one of the most important targets of selection. Despite the fundamental importance of this assumption (causative relationship between the load and fitness) it is still a hypothesis. The biggest problem is that the composition of the load is extremely complex and poorly known (thousands of SD mutations with different and very small individual phenotypic effects interacting with each other in additive or epistatic ways) and its effect on the fitness is noisy and hidden. In this project I would like to uncover a link between the mutational load and the fitness considering organisms under very severe conditions (hereafter the 'genetic handicap') where this predicted functional relationship between the load and viability (one of the most important components of fitness) can be stronger. I hypothesize that the severe conditions (genetic handicap) can interact with the load of SD variants in a specific way which makes the effect of each slightly-deleterious variant much more severe and thus much more detectable for selection. In my project I would like to test this statement from several points of view: theoretical (population genetic model and simulations), empirical (analyzes of live-born Down Syndrome individuals and carriers of rare deletions) and experimental (introduced genetic handicap into model species). Using the genetic handicap approach it will be possible to investigate deeply the mutation loads of human and model organisms. Happy new year and hope to see you there! Valentin (for the organizers: Jérôme Goudet, Valentin Rousson, Frédéric Schütz)

1 0

statistical forum on December 22 at 11:15
by Rousson Valentin 12 Dec '16

12 Dec '16

Dear all, The next seminar in our series of statistics talks for people interested in quantitative methods in biology and medicine will take place on Thursday 22 December 2016 at 11h15, in the Salle Delachaux (Biopôle 2, first floor, Route de la Corniche 10, 1010 Lausanne, M2: Vennes). The speaker will be Dr. Dominik Heinzmann (Associate Director Biostatistics, Roche, Basel, Switzerland) Title: Pharmacokinetic challenges and opportunities in drug development Abstract: The talk will first cover basic pharmacokinetic (PK) principles in drug development, including relevant research questions, endpoints to assess the objective, methods of measurement and interpretation. This general framework will then be discussed in the context of real case examples including the development of a different route of administration (subcutaneous versus intravenous) for oncology drugs, the development of an alternative device to administer drugs (handheld syringe versus a device) and the development of biosimilars (analogous to generics for biologics). Finally, feedback to the different approaches from various health authorities will be highlighted to provide the audience with a comprehensive overview on "viability" of those options. Hope to see you there! Valentin (for the organizers: Jérôme Goudet, Valentin Rousson, Frédéric Schütz)

1 0