Institute of Statistical Science Academia Sinica [Seminar Feed] http://www.stat.sinica.edu.tw Statistics, Stat, Edu en-us Fri, 17 Aug 2018 13:48:19 +0800 http://www.stat.sinica.edu.tw/statnewsite/seminar/rss/ PHP admin@stat.sinica.edu.tw admin@stat.sinica.edu.tw Introduction to deep learning and TensorFlow http://www.stat.sinica.edu.tw/statnewsite/seminar/show/2299/

Abstract

 

    With the advantages of big data, hardware, and algorithms, Artificial Intelligence (AI) began to flood people's lives. Deep Learning is one of the popular machine learning methods to implement AI. This talk will guide you to understand popular frameworks, models, and applications of deep learning. I will also introduce the TensorFlow, an open source software library for numerical computation using data flow graphs. This session will include some hands-on demonstration of the TensorFlow, it is welcome to bring a laptop and discuss together.

]]>
Tue, 7 Aug 2018 14:58:14 +0800 http://www.stat.sinica.edu.tw/statnewsite/seminar/show/2299/
Efficient variable selection to identify causal genetic variants and quantify their contribution to complex phenotypes http://www.stat.sinica.edu.tw/statnewsite/seminar/show/2296/

Abstract

 

    Genome-Wide Association Studies (GWAS) have narrowed down the genome into regions underlying complex phenotypes. However, any one region still harbours thousands of correlated genetic variants, complicating biological follow-up. We therefore need variable selection to refine the large set of variants simply associated with a phenotype down to a much smaller set of putative causal variants with direct effect on the phenotype. This is, however, a hard combinatorial problem and requires advanced statistical methods to efficiently explore the high-dimensional model space.

 

    We present the FINEMAP software that couples Bayesian variable selection for fine-mapping causal variants with an ultrafast high-resolution stochastic search. With extensive simulations we show that FINEMAP is as accurate as exhaustive search when the latter can be completed and achieves even higher accuracy when the latter must be constrained due to computational reasons. We further demonstrate that FINEMAP opens up completely new opportunities by fine-mapping the HDL-C association of the LIPC locus with 20,000 variants in less than 90 seconds while exhaustive search would require thousands of years.

 

    GWAS sample sizes soon counted in millions provide unprecedented opportunities for fast and accurate fine-mapping. It would further be useful to routinely evaluate how much of the phenotypic variation can be explained by the fine-mapped variants. Therefore, we compare regional heritability estimation using FINEMAP with both the variance component model BOLT and fixed-effect model HESS in 110 regions across 51 biomarkers on 5,265 Finns. Our results show good concordance among all methods in regions with negligible contribution to the genome-wide heritability, whereas BOLT and HESS yielded respectively larger and smaller estimates than FINEMAP in regions with moderate to high heritability levels. Scaling the analysis for lipid traits from 5,265 Finns to 21,320 Finns shows good agreement between FINEMAP and BOLT also for moderate to high levels of regional heritability, whereas HESS estimates are consistently lower at these levels. Through comprehensive simulations with biobank-scale projects, we illustrate how violations of model assumptions on polygenicity or unspecified genetic architecture induces inaccuracy to the existing heritability estimates that becomes more accentuated as statistical power to identify causal variants increases.

]]>
Fri, 13 Jul 2018 17:32:26 +0800 http://www.stat.sinica.edu.tw/statnewsite/seminar/show/2296/
Statistics with a human face http://www.stat.sinica.edu.tw/statnewsite/seminar/show/2301/

Abstract

    Three-dimensional surface imaging, through laser-scanning or stereo-photogrammetry, provides high-resolution data defining the surface shape of objects.  In an anatomical setting this can provide invaluable quantitative information, for example on the success of surgery.  Two particular applications are in the success of facial surgery and in developmental issues with associated facial shapes.  An initial challenge is to extract suitable information from these images, to characterise the surface shape in an informative manner.  Landmarks are traditionally used to good effect but these clearly do not adequately represent the very much richer information present in each digitised images. Curves with clear anatomical meaning provide a good compromise between informative representations of shape and simplicity of structure, as well as providing guiding information for full surface representations.  Some of the issues involved in analysing data of this type will be discussed and illustrated.  Modelling issues include the measurement of asymmetry and longitudinal patterns of growth.


]]>
Wed, 15 Aug 2018 09:13:20 +0800 http://www.stat.sinica.edu.tw/statnewsite/seminar/show/2301/