Xinyu Liu
*** ********* ** *** **, Athens, GA ***05. 706-***-****. **********@*****.***.
●QUALIFICAION
8+ years experience on (Generalized) Linear (Mixed) Model, Experimental Design, Machine Learning
Proficient in R, SAS, MATLAB, Python, Fortran, Linux with 6+ years experience
Excellent in survival analysis, sample size determination, and PK/P D modeling
Integrated background of Statistics, Bioinformatics, Biology
●EXPERIENCE
Intern at Dow AgroSciences through Kelly Services
Dow AgroSciences, Indianapolis May, 2012 -- Aug, 2012
Established a pipeline to automate the quality control and diagnosis for proteomics experiments
Investigated the statistical quality control methodology for proteomics data
Built an R package to evaluate and visualize the quality control metrics
Statistical Consulting Assistant
Statistical Consulting Center, University of Georgia 2010 -- Present
Conducted 15+ projects from different clients. Selected projects:
Conducted survival analysis on breast cancer data
Kaplan-Meier model and Cox proportional hazards model
This resulted in the journal article which has been recently submitted to Molecular Cancer
Estimated the sample size required by an investigation of co -teaching
Designed log-linear model and negative binomial model to analyze large scale genome data
Built predictive models to forecast desire for relocation in rural Georgia by housing characteristics
Support vector machine, neural network, decision tree, logistic regression
Designed GARCH model to test the effects of stock return from environment events
Research Assistant
Laboratory of T.N. Sriram, Department of Statistics, University of Georgia 2010 -- 2012
Have developed algorithms to estimate sample size needed for an SNP classifier (PhD Dissertation)
Built an R package to compute the sample size required by the SNP classifier
Analyzed a multiple-class ROC and VUS of SNP classifier
Laboratory of Animal and Dairy Science, University of Georgia 2008 -- 2009
Forecasted two intake-relative traits from low density SNP panel data (Cooperated with Sanofi Merial)
Developed Python package for SNP imputation and encoding
Designed algorithms to reduce the dimension of the SNP dataset
Forecasted the intake-relative traits from low density SNP penel data
Neural network, support vector machine, principle component analysis, cross validation
Developed an ant colony optimization algorithm to detect Epistatic Interactions for case -control studies
Xinyu Liu
100 Woodstone Dr APT 12, Athens, GA 30605. 706-***-****. **********@*****.***.
Research Assistant
Bioinformatics Department, University of Copenhagen 2005 --2007
Identified and Classified the repeats in non-coding regions of 58 Archaea genomes (Master Thesis)
Designed the algorithm to search the repetitive sequences in non -coding region
Developed a methodology to detect the novel mobile elements and non -coding RNA
Analyzed the Local Mass Fractal Dimension of 1520 Proteins
Designed an algorithm to compute the fractal property of proteins using Bio.PDB package
Forecasted ADMET according to the 2-D and 3-D descriptors of drugs by Machine Learning methods
Designed compartment PK/PD models for concentrations of digoxin in blood serum
Built non-linear mixed models by R
●EDUCATION
Ph.D Candidate in Statistics, University of Georgia, USA Available soon
M.S. in Statistics, University of Georgia, USA 2010
M.S. in Bioinformatics, University of Copenhagen, Denmark 2007
B.S. in Biological Sciences, Fudan University, China 2004
●CERTIFICAION
SAS Base Programming for SAS9 Dec, 2010
SAS Advanced Programming for SAS9 Jan, 2011
●PUBLICATION
X. Liu, Y. Wang, R. Rekaya, and T.N. Sriram. (2012) Sample size determination for classifiers based on
single nucleotide polymorphisms. Biostatistics, 13 (2), 217-227.
X. Liu, T.N. Sriram (2012). The ROC analysis and sample size determination of a multi -class SNP classifier.
Under Preparation.
Y. Wang, S. Joseph, X. Liu, M. Kelley and R. Rekaya. (2012) SNPxGE2: a database for human
SNP-coexpression associations. Bioinformatics, 28(3): 403-410.
Y. Wang, X. Liu, R Rekaya. (2012) AntEpiSeeker2. 0: extending epistasis detection to epistasis -associated
pathway inference using ant colony optimization. Nature Publishing Group 713: 1-10.
M. Davis, X. Liu, et al. An epigenetic regulator, Co-Activator Arginine Methyltransferase 1 (CARM1),
expression and sub-cellular localization is associated with specific breast cancer subtypes study. Submitted
to Molecular Cancer.
Y. Wang, X. Liu, M. Kelley, R. Rekaya. (2011) dMotifGreedy: a novel tool for de novo discovery of DN A
motifs with enhanced power of reporting distinct motifs. Arxiv preprint arXiv:1102.4015
Y. Wang, X. Liu, K. Robbins and R. Rekaya. (2010) AntEpiSeeker: detecting epistatic interactions for
case-control studies using a two-stage ant colony optimization algorithm. BMC Res Notes 3: 117.