- statslausanne - lists.sib.swiss

Symposium en l'honneur d'Alfio le 2 novembre
by Frederic Schutz 18 Oct '12

18 Oct '12

Dear list members, A scientific workshop will be held on *November 2 2012 at 14:00* (Friday) in honor of *Professor Alfio Marazzi* who has just retired from the University Hospital and from the University of Lausanne. _Workshop title:_ *STATISTICAL ANALYSIS OF HOSPITAL DATA AND THE ROLE OF ROBUST STATISTICS IN THIS REGARD* _Time_: Friday, November 2 2012 at 14:00 _Location_: Room Delachaux, first floor of building Biopôle 2, Route de la Corniche 10, 1010 Lausanne (metro station: Vennes) _Speakers_: * Olivier Peters and Hong Dung Duong (University Hospital Lausanne): "Les limites de la structure tarifaire SwissDRG - une méthode pour les contourner" * Peter Rückdeschel (Freihofer Institut, Kaiserslautern): "Bed at risk - how long must we stay robust(ly) ?" * Claudio Agostinelli (Ca' Foscari University, Venise): "Robust inference in Generalized Linear Models with applications to hospital costs" * Victor Yohai (Universidad de Buenos Aires): "Robust estimates of location and scatter matrix under the independent contamination model" An aperitif will be organized at the end of the workshop. If you intend to participate to the workshop and if you are not working at the University Hospital Lausanne, please send a short note to Catherine Turrian (catherine.turrian(a)chuv.ch) if possible before October 29, such that we can estimate the number of participants. We are looking forward to see you there! ---------------------------------------------------- Valentin Rousson Unité de Statistique Institut Universitaire de Médecine Sociale et Préventive Centre Hospitalier Universitaire et Université de Lausanne Biopôle 2 Route de la Corniche 10 CH-1010 Lausanne Tél : ++41 (021) 314 73 28 Fax : ++41 (021) 314 73 73 Email: Valentin.Rousson(a)chuv.ch <mailto:Valentin.Rousson@chuv.ch>

1 0

Next statistics forum: Darlene Goldstein and Alix Leboucq, Thursday 18 October 2012
by Frederic Schutz 15 Oct '12

15 Oct '12

Dear all, The next seminar in our series of statistics talks for people interested in quantitative methods in biology and medicine will take place Thursday 18 October at 11h00 in the Génopode Building in Dorigny (room 2020). The speakers will be Darlene Goldstein and Alix Leboucq from the EPFL, who will talk about "Combining microarray studies with missing information". Abstract: Meta-analysis of microarray studies to produce an overall gene list is relatively straightforward when complete data are available. When some studies are missing information (for example, some studies have only a ranked list of genes instead of complete primary data), a common approach is to reduce all studies to ranked lists prior to combining. Since this step entails a loss of information, we wondered whether we could improve the final list by attempting to preserve the information available for each study. We therefore consider a modeling approach to combine the studies using the type of information available in each study: full data matrix, summary statistics, or ranks for each gene. We will discuss progress in the development of the model, some results from a simulation study of model performance and from real data. You can find more information about the forum on our web page at http://moodle.unil.ch/course/view.php?id=2729 Hope to see you there ! Frédéric (for the organizers: Jérôme Goudet, Valentin Rousson, Frédéric Schütz)

1 0

Next statistics forum: Jan Beyersmann, Thursday 4 October 2012
by Frédéric Schütz 02 Oct '12

02 Oct '12

Dear all, The next seminar in our series of statistics talks for people interested in quantitative methods in biology and medicine will take place Thursday 4 October at 11h00 in the Salle Delachaux, Biopôle 2 (first floor), route de la Corniche 10, Lausanne (metro: M2, Vennes). The speaker will be Jan Beyersmann who will talk about "Competing risks and simulation". Abstract: The analysis of survival or time-to-event data is one of the most common applications of advanced statistical techniques in medical research and beyond. Data are typically incomplete as a consequence of limited observation periods. Therefore, survival analysis is based on hazards. Hazard-based techniques also allow for analysing competing risks, i.e., time-to-first-event and type-of-first-event. A common example from cancer research is relapse-free survival, which is the time until relapse or death, whatever comes first. A competing risks analysis distinguishes between these two event types, but the presence of more than one event-specific hazard poses a challenge both in theory and practice. We discuss the use of simulation to understand and analyse competing risks data, interpret the analyses and plan studies with competing risks outcomes. Reference: Beyersmann, J., Allignol, A., and Schumacher, M. (2012) Competing Risks and Multistate Models with R. Springer, New York You can find more information about the forum on our web page at http://moodle.unil.ch/course/view.php?id=2729 The next seminar will take place on 18 October in Dorigny. The speakers will be Darlene Goldstein and Alix Leboucq, from the EPFL. Hope to see you there ! Frédéric (for the organizers: Jérôme Goudet, Valentin Rousson, Frédéric Schütz)

1 0

Next statistics forum: Asa Wirapati, Thursday 20 September 2012
by Frederic Schutz 13 Sep '12

13 Sep '12

Dear all, we are starting again our series of statistics talks for people interested in quantitative methods in biology and medicine in the Lausanne community. Since is the first event of the series, please distribute this announcement widely; interested people can subscribe directly to the mailing-list from the page http://lists.isb-sib.ch/mailman/listinfo/statslausanne The next seminar will take place Thursday, 20 September at 11h00 in the Génopode Building in Dorigny (room 2020). The speaker will be Asa Wirapati, from the Bioinformatics Core Facility (SIB), who will talk about "Estimating the invisible: Effect size distributions in genome-wide association studies". The talk will be linked to the content of this paper: http://www.ncbi.nlm.nih.gov/pubmed/21118193 You can find more information about the forum on our web page at http://moodle.unil.ch/course/view.php?id=2729 Hope to see you there ! Frédéric (for the organizers: Jérôme Goudet, Valentin Rousson, Frédéric Schütz)

2 1

Next statistics forum: Zoltán Kutalik, Thursday 28 June, 11:00 in Dorigny
by Frederic Schutz 27 Jun '12

27 Jun '12

* Many people are not yet subscribed to the statslausanne mailing-list; please forward appropriately to your colleagues * Our next statistics forum for people interested in quantitative methods in biology and medicine will take place Thursday, 28 June at 11h00 in the Génopode Building in Dorigny (exact room still to be confirmed; there will be signs in any case once you enter the building). The speaker will be Zoltán Kutalik, from the Department of Medical Genetics, who will talk about "Estimating the invisible: Effect size distributions in genome-wide association studies". Abstract: Genome-wide association studies (GWAS) are conducted with the promise to discover novel genetic variants associated with diverse traits. For most traits, associated markers individually explain just a modest fraction of the phenotypic variation, but their number can well be in the hundreds. We developed a maximum likelihood method that allows us to infer the distribution of associated variants even when many of them were missed by chance. Compared to previous approaches, the novelty of our method is that it (a) does not require having an independent (unbiased) estimate of the effect sizes; (b) makes use of the complete distribution of P-values while allowing for the false discovery rate; (c) takes into account allelic heterogeneity and the SNP pruning strategy. We applied our method to the latest GWAS meta-analysis results of the GIANT consortium. It revealed that the observed P-values provide evidence for the existence of variants explaining several fold higher phenotypic variance than genome-wide significant SNPs. This methodology also enables us to predict the benefit of future GWA studies that aim to reveal more associated genetic markers via increased sample size. You can find more information about the forum on our web page at http://moodle.unil.ch/course/view.php?id=2729 Hope to see you there ! Frédéric

1 1

CANCELLED Next statistics forum: Darlene Goldstein and Alix Leboucq, this Thursday 24 May, 11:00 in Dorigny
by Frederic Schutz 22 May '12

22 May '12

We are sorry to announce that the statistics talk planned for this Thursday morning in Dorigny has to be cancelled. We'll keep you informed about the next talk, which will take place in June. Frédéric

1 0

Next statistics forum: Darlene Goldstein and Alix Leboucq, this Thursday 24 May, 11:00 in Dorigny
by Frédéric Schütz 21 May '12

21 May '12

* Many people are not yet subscribed to the statslausanne mailing-list; please forward appropriately to your colleagues * Our next statistics forum for people interested in quantitative methods in biology and medicine will take place this Thursday 24 May at 11h00 in the reading room (2020) of the Genopode building (2nd floor, same level as the main entrance) in Dorigny. The speakers will be Darlene Goldstein and Alix Leboucq from EPFL, who will talk about Combining microarray studies with missing information Abstract: Meta-analysis of microarray studies to produce an overall gene list is relatively straightforward when complete data are available. When some studies are missing information (for example, some studies have only a ranked list of genes instead of complete primary data), a common approach is to reduce all studies to ranked lists prior to combining. Since this step entails a loss of information, we wondered whether we could improve the final list by attempting to preserve the information available for each study. We therefore consider a modeling approach to combine the studies using the type of information available in each study: full data matrix, summary statistics, or ranks for each gene. We will discuss progress in the development of the model and some preliminary results from a simulation study of model performance. You can find more information about the forum on our web page at http://moodle.unil.ch/course/view.php?id=2729 Hope to see you there ! Frédéric

1 0

Next statistics forum: Alfio Marazzi, this Thursday 10 May, 11:00 in Epalinges
by Frederic Schutz 07 May '12

07 May '12

* Many people are not yet subscribed to the statslausanne mailing-list; please forward appropriately to your colleagues * * NOTE: for those of you who came to this meeting previously, the location has changed * Our next statistics forum for people interested in quantitative methods in biology and medicine will take place this Thursday, 10 May at 11h00 at the IUMSP, Biopôle 2, route de la Corniche 10, in the conference room (1st floor). The speaker will be Alfio Marazzi, from the IUMSP, who will talk about "The analysis of hospital cost and length of stays: robust estimation of the generalized loggamma model." Abstract: In the first non-technical part of my talk, I will motivate the use of robust statistics in the analysis of hospital length and cost of stay. Several methods for robust regression with asymmetric errors have been developed for this purpose. In the second technical part of the talk, I will briefly present a recently developed robust method to estimate the generalized loggamma model (a very flexible family of bsymmetric and asymmetric distributions, including normal, log-Weibull, and log-gamma). You can find more information about the forum on our web page at http://moodle.unil.ch/course/view.php?id=2729 Hope to see you there ! Frédéric

1 0

Next statistics forum: Jérôme Goudet, this Thursday 19 April, 11:00 in Dorigny
by Frédéric Schütz 17 Apr '12

17 Apr '12

* Many people are not yet subscribed to the statslausanne mailing-list; please forward appropriately to your colleagues * Our next statistics forum for people interested in quantitative methods in biology and medicine will take place this Thursday 19 April at 11h00 in the reading room (2020) of the Genopode building (2nd floor, same level as the entrance from the Amphimax building). The speaker will be Jérôme Goudet from UNIL, who will talk about "Partial least squares regression (PLS) versus patterned statistics in a Spatial Approximate Bayesian Computation (ABC) context" Abstract: Approximate Bayesian Computation (ABC, http://en.wikipedia.org/wiki/Approximate_Bayesian_computation ) is an increasingly used method to estimate parameters when obtaining the expression for the full likelihood is difficult or impossible. The idea is simple: it consists in simulating data under a process similar to that that generated the observation, and to compare the simulated and observed data via summary statistics. The simulations giving summary statistics closest to the observed are retained, and provide estimation of the though parameters. A key question is how many and what type of summary statistics should be used? In this presentation, I will illustrate the use of ABC for the estimation of demographic and genetic parameters during the colonization of Europe after the last glaciation by a bird of prey, the barn owl. Interestingly we find that a blind approach using several hundred summary statistics post processed via Partial Least Squares regression (PLS http://en.wikipedia.org/wiki/Partial_least_squares_regression ), gave essentially the same results as an analysis where we carefully picked 4 summary statistics that we thought would best describe the information contained in our data. You can find more information about the forum on our web page at http://moodle.unil.ch/course/view.php?id=2729 Hope to see you there ! Frédéric

1 0

Next statistics forum: this Thursday 5 April, 11:00 in Epalinges
by Frederic Schutz 02 Apr '12

02 Apr '12

* Many people are not yet subscribed to the statslausanne mailing-list; please forward appropriately to your colleagues * Our next statistics forum for people interested in quantitative methods in biology and medicine will take place this Thursday 5 April at 11h00 in the cafeteria, 2nd floor of the Biopole 1 building in Epalinges, The speaker will be Sven Bergmann from UNIL/SIB, who will talk about " Genome-wide association studies: Hype and reality " You can find more information about the forum on our web page at http://moodle.unil.ch/course/view.php?id=2729 Hope to see you there ! Frédéric

1 0