Current Offerings:
Large-scale Significance Testing of Genomic Data
John Storey (Princeton University)
July 23, noon - 2:00 p.m. Eastern time
The presenter will discuss recent advances in performing many hypothesis
tests in the context of genomics data. This will include discussion on the
false discovery rate, accounting for latent structure, and borrowing information
across variables to increase power.
Genomic Data Analysis with Targeted Maximum Likelihood and Super Learning
Mark van der Laan (UC Berkeley)
August 24, noon - 2:00 p.m. Eastern time
Current statistical practice to assess an effect of an intervention or exposure on an outcome of interest often involves either maximum likelihood estimation for a priori specified regression model, or, manual and/or data adaptive interventions to fine tune a choice of model. In both cases, bias in the point estimates and the estimate of the signal to noise ratio are rampant, causing an epidemic of false claims based on data analyses.
In this talk we present our efforts to construct machine learning algorithms for estimating a causal or adjusted effect that take away the need for specifying regression models, while still providing maximum likelihood based estimators and inference. Two fundamental concepts underlying this methodology are super learning, i.e., the very
aggressive use of cross-validation to select optimal combinations of many model fits, and subsequent targeted maximum likelihood estimation to target the fit towards the causal effect of interest. Our maximally unbiased and efficient estimates are accompanied with statistical inference. In addition, multiple testing methods are employed in case one pursues effect estimation across a large set of variables.
We illustrate this method in observational studies for assessing the effect of mutations in the HIV virus that cause resistance to a particular drug regimen. We also illustrate the performance for assessing the effect on the outcome or response to treatment of single
nucleotide polymorphisms and gene-expressions in genomic studies, including randomized trials. In particular, we demonstrate the performance of the super learning in prediction.
About Webinars
A webinar is a seminar which is conducted over the World Wide Web. It is a type of web conferencing. In contrast to a Webcast, which is transmission of information in one direction only, webinars are designed to be interactive between the presenter and audience. A webinar is 'live' in the sense that information is conveyed according to an agenda, with a starting and ending time. In the case of the Biopharmaceutical Section Webinar Series, the presenter speaks over a standard telephone line, pointing out information being presented on screen. The audience receives the audio of the presentation via telephone call-in number of audio streaming over the internet. The audience can respond via a chat feature. The word 'webinar' is a blend of web and seminar.
Attendee
User Guide
System Requirements
Operating System: Windows 2000 to present, Macintosh OSX, Linux Redhat
Browser: Internet Explorer (IE) 5.5+, Firefox 2.0+, Opera
9.0+, Mozilla 1.71+, Safari 3.1
Other: Events with streaming audio or video require Macromedia Flash 8.0+
Hardware: 56Kbps Internet access. Speakers or headphones and cable modem,
DSL, ISDN, or equivalent broadband needed to receive audio/video streaming
(128K minimum).
*Opera browser will not allow access to the Question and Answer feature
of the console.
More information on the Web-based training program can be found at the biopharmaceutical
network's web site.