Alison Motsinger-Reif

Alison Motsinger-Reif, PhD will lead the biostatistical components of the current project. She will collaborate on study design, power calculations, etc. during the first two years of the project, and will work on data analysis of the current project. She will dedicate, on average, a total of 0.6 calendar months for the first three years, and 1.2 months for the last three years. She will supervise a graduate student in developing and implementing analytical methods for the project.

This GIT will support Tao Jiang for Spring 2018.

A major challenge facing the AD research and medical community is that of diagnosing and treating a multifactorial disease based on incompletely understood pathophysiology and molecular mechanisms. While profound biochemical and pathological alterations have been reported involving abnormal amyloid beta metabolism, vascular changes, disruptions of neurotransmitters, oxidative stress and inflammatory processes, the relationships among these changes and the cascade of events leading to pathogenesis remains unclear. Drug trials targeting any one pathway in isolation (e.g., amyloid metabolism) have yet to identify effective therapies. A new perspective emerging from recent research is that AD is a network disorder involving a shift from normal to pathological networks and that a systems approach that defines biochemical connections is a critical first step to understanding disease heterogeneity and mechanisms of cognitive failure. ����������������Metabolomics,��������������� a new discipline committed to the union of biochemistry with quantitative systems biology and high-throughput techniques, provides powerful tools for mapping global biochemical changes in AD including failures in communication within biochemical pathways and metabolic networks. We have pioneered applications of metabolomics for the study of neuropsychiatric diseases including Alzheimer������������������s disease. Over four years we have assembled an interdisciplinary team of experts in metabolomics, genetics, biochemistry, bioinformatics, biomarker discovery and clinical trials and have begun to define perturbations in interlinked biochemical pathways across the trajectory of disease. In our initial studies, using non-targeted lipidomics platforms we identified changes in lipidome of AD patients that points to major down regulation of phospholipids including phosphatidylcholine (PC) plasmalogens and sphingolipidome. Using targeted and non-targeted metabolomics platforms we have identified major defects in the methionine (MET) /one carbon metabolism pathway ������������������ this pathway regulates fundamental cellular methylation and oxidative stress defense systems. It generates homocysteine whose elevation has been linked to neurotoxicity, oxidative stress, DNA damage, and increased risk for both stroke and dementia. Perturbation in norepinephrine (NE), tryptophan (TRP) and purine (PUR) pathways were also identified reflecting failures in neurotransmission and mitochondrial dysfunction and these too connect with the MET pathway. Constructed metabolic networks linked perturbations in NE and PUR with elevated tau and changes in TRP and MET to amyloid-beta42. Leveraging large NIH investments in the Alzheimer Disease Neuroimaging Initiative (ADNI), Pharmacometabolomics Research Network, Murdock community study, we have started to link metabolomic findings with vast genetics and imaging data sets and begun building an integrated systems approach to the study of AD.

In addition to high blood sugar (hyperglycemia), diabetic patients frequently have high blood pressure (hypertension) and high levels of LDL-cholesterol and triglycerides, often coupled with low HDL-cholesterol (dyslipidemia).The majority of diabetes related mortality is due to cardiovascular events, and epidemiological studies have shown that cardiovascular risk increases with increasing levels of blood sugar, blood pressure, and blood lipids. A variety of drugs are available to treat each of these conditions, and some have been shown to have an effect on cardiovascular risk. For example, controlling LDL-cholesterol with statin therapy reduces the rate of cardiovascular events in diabetic patients, but not to the level characteristic of non-diabetic individuals. The ACCORD trial investigated whether intensive pharmacological therapy in diabetic patients, with the goal of normalizing glycemia, blood pressure, and blood lipids, would further reduce cardiovascular events. However, no additional effect was seen with intensive blood pressure or lipid therapy, and intensive glycemia management actually increased mortality. These failures of seemingly rational treatment approaches could be the result of differential response of individuals to particular therapeutic regimens due to genetic polymorphism in genes relating to the metabolism or mechanism of action of the medicines used. Many candidate genes could be advanced as possible sources of this genetic variation, but our knowledge of all genes contributing to metabolic and cardiovascular phenotypes is incomplete, and therefore a candidate gene approach cannot be assured of identifying the relevant genes. We therefore propose a genetic study of the ACCORD trial that looks at functionally significant genetic variation in all genes in the human genome to investigate the following questions: 1) Does genetic variation between individuals explain differences in response to fenofibrate, a drug that lowers triglycerides and raises HDL? 2) Does genetic variation between individuals explain differences in response to statins, drugs that lower LDL? 3) Does genetic variation between individuals explain differences in response to individual anti-hyperglycemia drugs in the short term, and to the intensive therapy approach to glycemia management in the longer term? Identification of genetic variants affecting outcomes of glycemia and lipid modifying therapies would enable the targeting of particular interventions to patients most likely to benefit and least likely to be harmed, improving cardiovascular outcomes and reducing the burden of morbidity and mortality attributable to diabetes. The genes containing these variants may prove to be novel targets for drug development, leading to new medicines for improving outcomes for diabetic patients in the future.

Siamak Mahmoudian Dehkordi will work on software pipeline development and refinement working with Vivitras Therapeutics. He will dedicate 20 hours per week working on their project. This will entail software development and methods comparisons, focused on determining secondary structure from sequencing data.

The mortality rate in immune-mediated hemolytic anemia (IMHA) in the dog is reported to be as high as 70%, and over 50% of these deaths are attributed to thromboembolism. Thus, a readily accessible and accurate method to monitor anticoagulant therapy is urgently needed to improve survival. In a recent on-line survey of 27 United States veterinary colleges performed by the PI, 95% of the 19 respondents reported their institution used unfractionated heparin (UFH), with or without antiplatelet therapy, for thromboprophylaxis in hospitalized dogs with IMHA. Most (70%) survey respondents monitored UFH therapy using activated partial thromboplastin times (aPTT). Although aPTT is easy to use, variations in reagents and instrumentation create significant disparities in its measured response. Measurement of the amidolytic activity of the plasma protease Factor Xa, which is blocked by UFH, is considered the gold standard for heparin monitoring. Yet, our survey indicates that few veterinary colleges (3/19) had the ability to perform this assay and only 1 reports its routine use. The assay is complex, with inherent background activity, and is not cost effective without running a large number of samples, making its use unlikely to become widespread or available cageside. The survey also revealed that most (79%) respondents had access to in-house thromboelastography (TEG). Our long-term goal is to utilize a whole blood viscoelastic-based assay, such as TEG, to provide objective point-of-care measurements of clot kinetics and strength to more effectively titrate thromboprophylaxis with UFH therapy. An increased understanding of the importance of TEG monitoring for managing UFH therapy in IMHA patients is expected to result in an immediate improvement in survival. This cost effective diagnostic aide is likely to gain widespread use in the veterinary referral setting, fostering the dissemination of information to general practitioners

Diseases linked to Metabolic Syndrome (MetS) such at type-2 diabetes and cardiovascular disease are rapidly increasing due to the influences of a modern Westernized-life style, but the genetic, environmental, and physiological mechanisms linking the symptoms of Metabolic-syndrome remain to be elucidated. Large scale studies to systematically assess the how genotype interacts with the environment to cause complex disease are very difficult in humans, but such studies are relatively tractable in genetic models systems such as Drosophila melanogaster. We have shown previously that that there is a very substantial contribution of genotype-by-environment interactions to the phenotypic variation observed for MetS-like symptoms in a naturally genetically variable population of D. melanogaster. We have also been able to demonstrate clear correlations between metabolomic and gene expression profiles and these symptoms as they vary across diet. Finally, we have shown that genetic variance in some of these traits increase with a perturbing high fat diet, indicating the exposure of cryptic genetic variation for these symptoms could contribute to increases in disease. In this study we will build off the community resources for complex genetic trait analysis of the Macdonald-Long synthetic recombinant inbred line (RIL) population and the Drosophila Genomic Resource Population (DGRP) to map the genetic basis of genotype-by-diet interactions. First, using the 1700 Macdonald-Long Advanced Intercross synthetic RILs, we will map the genetic basis of MetS-like symptoms and the regions controlling genotype-by-environment interactions contributing to these symptoms to within 1 cM of the causal locus when the flies are raised on a ?normal? verses ?high fat? diet. We should be able to estimate both the effect size and population frequency of causative alleles. Second, based of the phenotypes measured in the F1 RIL population, 200 lines demonstrating the largest genotype-by-diet interaction effects will be selected for metabolomic and expression profiling. Metabolomic profiling will identify several hundred primary metabolite and whole genome expression profiles will be generated by microarray analysis. We will characterize the metabolomic and expression module structure that drives the genotype-by-environment interactions and link those pathways back to specific genetic variants. Finally, we will attempt to replicate the findings from the synthetic RIL population through association mapping in the natural variants represented in the 192 lines of the DGRP. The ultimate goal of this work is to identify genomic regions, metabolic pathways, physiological mechanisms, and dietary influences likely to be of importance to Metabolic Syndrome in humans.

Rationale: Osteosarcoma (OS) accounts for 85% of all primary bone tumors in dogs. OS is generally a highly aggressive cancer, with metastases to lung, or euthanasia, reported as the most common cause of death. With standard of care treatment, outcome is highly variable; ~25% of OS patients will survive for <3 months, 50% of OS patients will survive for about a year, and only ~20% will survive for >two years. We have shown in other canine cancers that DNA copy number aberrations (CNAs) are associated significantly with the duration of disease free interval (DFI) in canine patients treated with specific therapies. In this study we will use existing samples from OS patients enrolled in a clinical trial, to identify CNAs associated with DFI. Overall aim is to develop a molecular test to predict duration of DFI in OS patients treated with standard of care therapy. Hypothesis/Objectives: The hypothesis for this study is that DNA copy number aberrations in canine appendicular OS are significantly associated with duration of DFI when the patient is treated with standard of care (SOC) amputation and chemotherapy. The primary objective of this study is to perform high-resolution genome-wide assessment of DNA copy number aberrations (CNAs), verified by multicolor fluorescence in situ hybridization (FISH), using a cohort of canine appendicular OS patients that SOC surgery and chemotherapy. With known duration of DFI, this study will identify those CNAs that separate tumors (and hence patients) into prognostic outcome groups. These data will be used to generate an accessible cytogenetic prognostic test that may be used to aid OS patient management decisions. Study Design: The study population comprises OS tumor samples from 84 patients, 47 from a published clinical trial and 37 additional cases treated with the same clinical trial protocol. All cases were treated at a single site and are associated with full clinical workup, including time to detection of metastases and overall survival. The study is powered to identify those CNAs that can separate cases into prognostic groups. This study will use high-resolution genome-wide DNA copy number profiling of tumor DNA samples to identify CNAs in each case. The copy number changes will be verified by multicolor FISH analysis using high throughout multi-plane imaging of tumor biopsy specimens. Data will be analyzed with a two-stage approach; detect associations using extreme value sampling, and then assessment of associated CNAs in the full cohort to obtain accurate estimates of the effect size. This approach maximizes efficiency of the overall study. Preliminary Data: High-resolution genome-wide DNA copy number profiling of over 130 OS samples indicated that there were no major differences in the distribution of copy number aberrations (CNAs) across dog breeds. Analysis of the CNAs in a subset of the study cohort indicated substantial differences in the presence of CNAs in those dogs in the 25th and 75th percentiles of DFI. In the pilot set, six regions of the dog genome presented with CNAs unique to each of the two DFI group. These CNAs were verified by FISH analysis in diagnostic biopsy specimens. These data suggest we are already heading towards identifying regions of the canine genome where differential incidence of CNAs will segregate canine OS patients according to their likelihood of duration of DFI when treated with SOC. Importantly, each of these six aberrations has been observed previously in CGH profiles of OS tumor DNA samples isolated from numerous OS cases out with the current study cohort. While we do not know the DFI/outcome of these particular previous cases, the fact that they share CNAs indicates that the aberrations are valid across numerous breeds. Identification of such DFI-associated aberrations will highlight genes of interest that may be pursued in subsequent studies with a view to development of new therapies. Expected Results: The expected outcome of this project is the development of an accessible molecular cytogenetic assay to predict DFI in canine OS

A need exists in both the food animal industry and academia for professionals who can apply methodologies from statistics, quantitative genetics, and traditional animal breeding to food animal improvement. In addition, the Agriculture and Food Research Initiative Competitive Grants Program has identified Food Availability as a critical factor for ensuring Global Food Security. Genetic improvement is a critical part of ensuring food availability. Many new genomic tools which may be applied to food animal genetic improvement are now available. Application of those genomic tools to food animal genetics requires an in-depth understanding of statistics, quantitative genetics, and traditional animal breeding. The Department of Animal Science at North Carolina State University has a long history of training the types of professionals which will fulfill the described need. Our Overall Goal is tol recruit, train, and prepare professionals for successful careers in food animal genetics. Specifically we will: 1) Recruit, train, and prepare two doctoral students for successful careers in food animal genetics; 2) Recruit, train, and prepare two masters students for successful careers in food animal genetics; 3) Develop recruitment tools to attract students from traditionally underrepresented groups. Students will: 1) complete cross-disciplinary curriculum integrating quantitative genetics, statistics, and animal breeding; 2) interact with adjunct faculty members currently employed in the food animal industry; 3) develop, complete, and publish novel research with direct industry relevance; 4) complete course work in personal development and ethics; 5) attain relevant industry experience through completion of a cooperative internship with an industry partner. This proposal will prepare 4 professionals for careers in food animal genetics The outcome will be an increase in the number of professionals capable of apply statistical, quantitative genetic and traditional animal breeding methods to food animal improvement.

This program project, entitled ?Statistical Methods for Cancer Clinical Trials,? will be joint venture of Duke University, North Carolina State University (NCSU), and the University of North Carolina at Chapel Hill (UNC). Biostatistician and clinician researchers from these three top research institutions will collaborate on project research and share project-related resources. The scientific goal of this ambitious program project is to develop innovative statistical methods for cancer clinical trials that can hasten successful introduction of effective new therapies into practice. The method of approach is to leverage recent advances in statistical and computational science to create new clinical trial designs and data analysis tools that resolve many of the key scientific limitations of current clinical trial methodology. The program project involves five interrelated research projects focusing on practical design and analysis problems in Phase II and III clinical trials, the problem of missing data and efficient use of prognostic information, post-marketing surveillance and comparative effectiveness research using clinical trial data, pharmacogenetics and individualized therapies, and the potential of dynamic treatment regimes to improve cancer treatment. The proposed methods have the potential to alter the prevailing clinical trial paradigm and increase the discovery and translation of new treatments into clinical practice. The multi-institutional approach, which exploits the complementary strengths of each the three universities, includes an effective and energetic process for coordinated implementation, communication, and dissemination of the results, including development of new software for public dissemination to practitioners. The project will lead to significant improvements in cancer clinical trial practice that will result in improved health and outcomes for cancer patients The NCSU component of the project will be administered by the NCSU Center for Quantitative Sciences in Biomedicine (CQSB) and involves seven faculty in the Department of Statistics, who will participate in all facets of the research. The investigators will collaborate with each other and investigators from Duke and UNC to address the research problems jointly through synergistic interactions that exploit the complementary expertise from all three institutions.