Confidential, Ridgefield, CT
- Developed Web applications in R and R Shiny to automated statistical analysis for clinical drug development. The application includes clinical trial reporting, interactive graphics to present data at desired level per CDISC standards and guidelines, analyze data categories per user request, such as patient demography, drug efficacy, lab results, and adverse events. The application was designed to identify and use the appropriate data sources for analysis, such subject levels, AE, and SDTM.
- Created a web - based application for treatment applications, which serves as an interactive tool for assigning treatments and cohorts with built-in rules to enforce data accuracy and integrity. It also generates reports and emails to audit and summarize treatment allocations.
- Worked with plant researchers to discover new compound targets for crop protection & improved yield.
- Used R to illustrate disease progress over time for different systems and gain a better understanding of factors and interactions to cause crop disease.
- Created and administered user accounts on shiny server pro for users and testers and their privileges.
- Calculated and displayed the area under the disease progress curve (AUDPC) as a user tool to compare disease severity and quantitative summary of disease intensity over time using trapezoidal method.
- Extensively used R packages for data management and visualizations, such as dplyr, ggplot2, and shiny.
- Enhanced field trial models and discovery models using R packages for MLE and mixed models.
- Developed R algorithms to fit regression and nonlinear mixed effect models to environmental study datasets.
Confidential, Springfield, VA
- Developed and maintained SAS programs for Informed Visibility (IV) system as the new Confidential real-time, single source to manage all mail information, and estimation routines on mail processing data.
- Performed hypothesis testing of significant differences of end-to-end mail on-time scores and service variances between legacy and new mail performance measurement systems.
- Generated and automated quarterly mail volume and trending reports using SAS and ORACLE.
- Developed algorithms to analyze impacts of first/last mile profiles on scanned mail sampling, such as simulation and random sampling routines.
- Performed quality control checks for parallel testing of new and legacy performance measure systems and presented results as summary tables, reports and visualizations.
- Identified the discrepancies between data sets generated by two different systems using SAS data steps and procs, such as merge, summary, compare, transpose, etc., and investigated causes of discrepancies such as rounding, missing values, database refresh dates, etc.
Confidential, Lawrenceville, NJ
- Analyzed and reviewed adhoc SAS programs for conversion into R and R Shiny applications.
- Design web apps in R and R shiny for Exploratory Data Analysis of Biomarkers for precision medicine initiative.
- The custom web interface provides the functionality to allow researchers to visualize and evaluate biomarkers, endpoints and interactions on patient data.
- Designed and presented statistical results with R packages to perform data modeling, data mining, biomarker prediction and monitoring.
- Used SVN for R package version control, and Domino server to host R Shiny applications.
- Collected, analyzed and provided information for Clinical Trial Teams for decision-making.
- Built R statistical models such as multivariate and generalized linear models for analyzing, predicting, and visualizing the patterns, associations, and relationships among biomarkers.
Confidential, Arlington, TX
- Used Python to create web applications on servers, connect to databases and manage financial files.
- Worked with business analysts to understand the business processes and goals, prepare data, design, evaluate and deployment data analytic models.
- Developed procedures to consolidate data from various General Ledger and Payroll sources to allow users to validate and reconcile data between GL and payroll.
- Worked with database systems, such as NoSQL and MongoDB to analyze financial information.