Implementing Appropriate Multivariate Methods for Higher Quality Results from Gen

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): Currently, there is a distinct shortage of (bio) statistically proficient researchers and appropriate statistical tools designed by, and especially for, research in the genetics of substance abuse. This is underscored by the recent NIDA executive summary that expressed caution on the interpretation of genome-wide-and more generally, genetics-studies for substance abuse populations (National Institute on Drug Abuse [NIDA], 2010). The report strongly recommended that NIDA place its resources towards an "increase [in] support for training in biostatistics, statistical genetics, and related bioinformatcs disciplines-key support disciplines in which personnel shortages are already a constraint." (NIDA, 2010, p. iv). With this proposal, I aim to decrease both the 1) "personnel shortage" by receiving and providing appropriate training and 2) lack of appropriate tools for GWAS in substance abuse research. Recently appeared and hailed as a "breakthrough" by the journal Science (Pennisi, 2007) Genome Wide Association Studies (GWAS) are already used in many fields. For example, a recent search resulted in number of studies between 1,008 (www.gwascentral.org) to over 26,100 (Google Scholar) GWAS results. GWAS have the potential to shed light on the genetic contributions of complex disorders and elucidate the biological systems involved in their etiology, but only with valid methodological approaches. But despite such a range of applications, GWAS have yet to yield any "breakthrough" results, especially in substance abuse (Hutchison, 2010). Additionally, the growth in the number of GWAS-especially in alcohol and drug addiction research (Hutchinson, 2010; Loth, Carvalho & Schumann, 2011)-is not paralleled by a growth of rigorous statistical methods or tools that could fully exploit these complex data sets. This lack of breakthroughs in GWAS for substance abuse may be due to the lack of breakthroughs in appropriate statistical techniques. Statistical tool development could greatly facilitate new discoveries in the genetics of substance abuse and dependence, such as novel genetic markers, environmental influences and endophenotypes. In this proposal I have two primary scientific aims: 1) develop and make available appropriate tools (freely) for GWAS in substance abuse and 2) apply appropriate tools in order to elucidate the genetic and environmental influences of impulsive behaviors and substance abuse (in marijuana, nicotine and binge eating populations). Current preliminary results show both 1) replication of genetic markers and 2) novel genetic markers for substance abuse. Preliminary results show neuropeptide Y (NPY) markers are more associated to binge eating and polysubstance users than any other group. Additionally, I have identified several possibly protective markers related to the BDNF and MAOA genes.