Contractor to US FDA (formerly with Midnight Sun Technologies, later with Koniag Government Services), in the Quantitative Clinical Pharmacology group, in Division of Quantitative Methods and Modeling (DQMM) | Office of Research and Standards (ORS) | Office of Generic Drugs (OGD) | Center for Drug Evaluation and Research (CDER) | U.S. Food and Drug Administration (FDA)).

I was involved in validation for KIWI, Cognigen's proprietary database for NONMEM output.

With collaborators, developed novel approaches for flow cytometry data analysis, both for cases involving a target population (GAFF) and for exploratory analyses (SWIFT). SWIFT can identify extremely rare cell populations and facilitates more rigorous inferences than existing common practice. Publication DOIs: 10.1109/ICASSP.2010.5495653; 10.1002/cyto.a.22446; and 10.1002/cyto.a.22445.Investigated various data transformations for flow cytometry data.Designed an event-level database in PostgreSQL intended to integrate SWIFT, GAFF, and other programs to facilitate consensus clustering and theoretically justified multi-model inferences while providing additional convenience and visualization functionality. Presented poster at ISAC 2013.Initiated close collaboration with Prof. Gregor von Laszewski and served on the educational committee of his masters student, Andrew Pangborn, resulting in the parallelization of the EM algorithm using Nvidia's CUDA technology. Publication DOI: 10.1109/ISPA.2009.29.Co-wrote (with Prof. von Laszewski) two NSF CDI (Cyber-Enabled Discovery and Innovation) grant proposals. These got excellent scores both years.Helped write several NIH grants for principal investigator (Tim R. Mosmann).Made an Access relational database (~ 100 tables) for all Luminex data (XML output) for a multi-year rheumatoid arthritis clinical study, then used R for exploratory analysis for differences between groups.Did various statistical analyses of T cell immunology data in R, and data cleaning in PowerShell.Used logistic regression in a pilot study to show that the best linear combination of laboratory predictors of opportunistic infections in HIV+ patients who may otherwise have normal CD4 counts was RSV net spots (ELISpot assay), background spots, and CD8 T cell count.

Modeled viral growth dynamics in culture using Berkeley Madonna. DOI: 10.1007/s11538-008-9323-4.Participated in other Center for Biodefense Immune Modeling activities, and other research and statistics courses as required by my supervisor (David Oakes) and his training grant; see Education section below.

Made an improved model for JAK-STAT signal transduction after IFN-γ ligand binding using our in-house software BioNetGen, with Drs. James Faeder and William Hlavacek in the Cell Signaling Group. BioNetGen uses rule based modeling to handle the combinatorial complexity arising from multi-subunit stoichiometry. Presented poster at 6th International Conference on Systems Biology (ICSB 2005).Developed an ODE model of lymphocyte and viral dynamics to explain the absence of pathology despite high viremia in SIV-infected sooty mangabeys, with Drs. Ruy Ribiero and Alan Perelson.

(This was exactly a continuation of my previous post; I'm splitting it up here for technical accuracy.)I continued to work closely with the data management group on a sophisticated relational database, and did the majority of the data analysis.Publication DOIs: 10.1002/cyto.b.20146 and 10.1371/journal.pone.0014626.

Although I was a postdoc for Oxford University, I lived and worked at MRC Laboratories, Fajara, The Gambia as a long term visitor. (I'm splitting it up here for technical accuracy.)Through Oxford University's Nuffield Department of Clinical Medicine, within Prof. Adrian Hill's group, under Dr. Samuel J. McConkey, I worked on a phase II clinical trial investigating a DNA, MVA prime, boost therapeutic vaccine for chronic hepatitis B virus infection. I performed intracellular cytokine staining assays using a four-color FACSCalibur™ flow cytometer, supervised technicians doing IFN-γ ELISpot assays in a biosafety level 2 laboratory, and worked closely with the data management group on a sophisticated relational database.