Institute of Statistical Science Academia Sinica [Seminar Feed] Statistics, Stat, Edu en-us Fri, 20 Oct 2017 18:42:41 +0800 PHP Harnessing Omics Data to Dissect Complex Traits, Drug Responses, and Population Ancestry Abstract

With rapid advances in high-throughput and high-dimensional molecular profiling in recent years, a wealth of omics data becomes available. Harnessing omics data presents unmet opportunities to decipher the mechanisms of complex diseases, drug responses, and population ancestry. In this talk, I will present our recent accomplishments in developing biostatistical and bioinformatics methodologies and analysis tools for analyzing genomics, transcriptomics, epigenomics, and metabolomics data. I will also present our biological findings in studying population genomics and pharmacogenomics/pharmacoepigenomics.

Wed, 11 Oct 2017 14:32:37 +0800
The environmental data deluge-sinking or swimming? Abstract

    Technological advances in sensing, data acquisition, and mobile devices are leading to increasing amounts of environmental data being sampled more rapidly and comprehensively than ever before, and environmental monitoring is benefiting from these advances. If we are to acquire novel insights and knowledge from these data, it needs to be matched by innovations in data management, storage and retrieval and ultimately in data analytics and inference.  This is the world of knowledge creation and data-driven discovery. The many forms of environmental data, their complexity and variations present challenges, including making model-based inferences about patterns in the presence of uncertainty which then need to be communicated and visualized for different audiences.

Tue, 3 Oct 2017 09:24:10 +0800
The Global Young Academy: equipping scientists across disciplines to impact society Abstract

    In this talk, I will talk about the changing challenges we face in society over time, the role that science has played, and the importance of a new cadre of scientists who can work across disciplines, collaborating to tackle complex societal challenges, and the changing skills sets required. I'll then touch on the role GYA is playing in addressing this need for early and mid-career scientists, and finish with some key examples of our flagship projects and relevant project examples. We can also discuss how the GYA could support establishment of a national young academy in Taiwan.

Thu, 19 Oct 2017 15:56:25 +0800
Heterozygosity mapping for dominant trait variants Abstract

    Homozygosity Mapping (HM) is a well-known technique to identify long runs of genetic marker homozygosity that are likely to harbor genes responsible for recessive disease, but there has been no comparable method for dominant traits. We have developed an approach to map dominant disease genes based on heterozygosity frequencies of sequence variants in the immediate vicinity of a dominant trait. Based on theoretical analysis, computer simulation, and sequence data in patients with known pathogenic variants, the new Heterozygosity Analysis (HA) represents a powerful tool for finding dominant disease variants. In contrast to HM (based on recombination), our HA method relies on linkage disequilibrium, which dissipates rapidly with increasing distance from a trait variant, so that HA tends to identify dominant disease variants rather accurately.

    As part of our statistical analysis, we developed a novel approach to test for homogeneity, assuming heterogeneity as the null hypothesis (equivalence testing), which appears the most plausible situation for many complex traits.

Tue, 3 Oct 2017 09:49:29 +0800