• Implement a clinical document classifier using Python library Pandas, TextBlog, and machine learning algorithm in scikit-learn in order to incorporate into hospital natural language processing (NLP) pipeline with 80% accuracy• Function as SME to facilitate HL7 lab results parsers in Hadoop which reduce 70% of query time and experiment with machine learning algorithms using MLlib to fine-tune and study potential factors which may influence patient length of stay (LOS)• Apply SQL query to extract patient information in Cerner Millennium (Oracle database) and R package, dplyr, to perform data manipulation and quantitative analysis to improve system information policies and patient care• Design lab system workflows with medical directors, lab managers to be cost-effective and compliant to regulatory standards and provide 7 NYPH campuses with daily application level support and off-hour issue resolution for laboratory information systems

• Led implementation, supported, and maintained of a new next generation sequencing lab information system called GenoLogics•Took initiative to build Python scripts leveraging RESTfulAPI, which reduced reagent calculation time 60-fold • Improved and developed CLIA certified panels in Pharmacogenomics and Cancer for clinical use (NGS platforms)• Designed and implemented PostgreSQL queries and combined with software Jasper Report to generate lab audit reports

• Developed a genome-wide screening approach using Saccharomyces cerevisiae to identify and characterize resistance mechanisms against the volatile compound 1-octen-3-ol • Analyzed gene interactions using Ospray software, Funspec gene ontology cluster program, and SGD data base.

• Transformed yeast deletion library with plasmids harboring protein toxin genes (ricin, stx1,stx2, and PAP). Identified resistant strains and the associated knocked out genes were analyzed with bioinformatic tools.• In vitro translation screen for small molecules that inhibit ricin and stx2.

• Utilized BasePlate (TAP), Tecan Genesis 200, Matrix-PlateMate, CyBio-well, Zymark plate labeler and REMP heat sealer for high- throughput chemical screening• Analyzed multichannel pipetting machine output data and accuracy optimization for quality control