biostatistics, clinical trial, omics, biology

Research Statement
My research activities, both theoretical and methodological as well as my collaborative and applied work, are focused on addressing data aspects of biological systems and biostatistics. An inventive impact of my research activities is the creation of a fresh perspective to biological data analysis where a research assumes that there are logical stochastic processes that generate underlying population trajectories that can be observed at distinct time points with some measurement error. This new perspective creates a new school of thought for biostatistics estimation techniques that account for the noted correlations. The perspective that efficient data manipulation procedures should directly target covariance structures inspires the estimation techniques. I adhere to this school of thought by creating estimators that are based on new regularity measures on functional covariance induced by population trajectories. I developed this estimation philosophy for biological data manipulation in response to the large volumes of functional data that I encountered in my practical projects and to address a scarcity of coalescing logical justification for the analysis of functional data that accounts for covariance. Applying this framework has allowed me to produce potentially powerful and interesting results. I am also involved in collaborative and applied biological data management projects dealing with a range of design structures within scientific settings.
Practicum: Phase II Basket Group Sequential Clinical Trial with Binary Responses, from December 2015 to the present: Given that the value of biological data is only as good as the processes that produce that data, developing and applying well documented biological research procedures play an important part in research practices.

In collaboration with Ao Yuan and Ming Tan from the Department of Biostatistics, Bioinformatics, and Biomathematics at Georgetown University as my mentor, I used R to do simulation studies and performance evaluations, adapting Lan and DeMets’ method to control type I error. Also, I used frailty model to account for dependence and developed a framework for Phase II basket group sequential design with binary responses. These methodologies were motivated by the desire to produce functional results that were of value in clinical practice.
Position as a student research assistant, from September 2015 to the present: I have been working with Hongbin Fang and Xiaogang Zhong as my supervisors in the Department of Biostatistics, Bioinformatics, and Biomathematics at Georgetown University. The work entailed applying multivariate logistic regression models, analyzing large dimensional data sets, conducting statistical analyses using appropriate computation and graphical software, and developing the next generation sequencing data processing, analysis, and reporting pipeline. Engaging in the activities expected of my position as a student research assistant aid me in gaining valuable skills, knowledge, and experience.
Data Analysis and Reporting Experience, from August 2016 to the present: I have been working with Kepher Makambi as my instructor. The work entails comparing the distribution of vitamin C between Diabetics and Non-Diabetics in the US, identifying the effect of smoking status on steroid hormone levels. In addition, I apply linear regression model to determine factors influencing infant’s birth weight, determining the link between vigorous physical activity and breast cancer risk in Black women within the US, and identify the factors associated with PFS and OS among CTC groups. The position has allowed me to gain practical knowledge on how to handle biological data.
Other collaborative projects: In addition, I have worked with other mentors to complete collaborative projects in the Department of Animal Biotechnology at Shanghai Jiao Tong University. Firstly, I prepared my thesis on Prediction of Heterosis Based on Whole-genome SNPs and Trait-specific SNPs between June 2014 and June 2015. The activities entailed using R and Perl to develop pipeline to get trait-specific SNPs from QTLdb. Also, I wrote R functions to build kinship and SDAF to predict heterosis among breeds. Second, I worked with Yan You as my mentor from October 2012 to December 2013 on a National Innovative Practice Project titled Preparation of Anti-human Midkine Monoclonal Antibody. The activities setting up different immune procedures and different cell fusion methods to prepare hybridoma cell, and using limiting dilution method and ELISA for screening mouse cell lines secreting anti-human Midkine mAb.
Future work: A powerful and interesting use of the frameworks I developed for biological data analysis is the development of methods for conducting a functional classification and discriminate analysis. In addition, I am interested in extending the data analysis tools I developed to functional clinical trials. All my theoretical and methodological work is extendable to any situation involving biological data manipulation, and I intend on extending them to microarrays, proteomics, and metabolomics. In addition to the practical projects, I am interested in continuing collaborations with other research. With my academic background in biotechnology and biostatistics, I am well suited to engage in research involving the analysis of large dimensional data sets such as microarray, proteomics, and metabolomics data for clinical or basic research purposes. I have further experience and training in software use to include R, SAS, Perl, Python, C, C++, LaTeX, LyX, Linux, Microsoft Office, SPSS, MATLAB, and Mathematica. My intention is to have the capacity to solve open biological problems to which I can apply
biostatistics and biotechnology skills in the development of useful and novel statistical methodologies.