● Conceptualized and designed a CRISPR-Cas12a library of guide RNAs, with genomic, transcriptomic, and epigenomic data to ensure comprehensive coverage.

● Investigated drivers of CRISPR efficacy, using ML, resulting in a 15% increase in targeting accuracy.

● Led computational analyses of multiple CRISPR screens in rare diseases, identifying 400 gene hits.

● Co-authored manuscript on key regulators of human stem cell differentiation.

● Collaborated with wet-lab scientists to support 5 projects, accelerating project timelines by up to 20%.

● Conducted comparative analyses of off-target detection methods in CRISPR screens using single-cell data, employing statistical models to evaluate accuracy.

● Developed a custom sequence alignment pipeline for gene editing, improving alignment speed by 80%.

● Automated workflows for sequence alignment, NGS data analysis, and off-target detection, reducing processing time by 60% and enhancing collaboration with experimental teams.

● Implemented company’s first version control processes, standardizing collaborative software development and data analysis.

● Optimized clinical trials by streamlining biomarker analysis using ML models, including regression, random forest, and Xgboost, as well as explainable AI.

● Improved sample quality by 10% across 49 clinical trial arms with department’s first ML pipeline to study the effects of biomarker operations on the quality of cancer immunotherapy assays.

● Upgraded data operations for 6 data streams, including flow cytometry, biomarker operations, and oncology data warehouses, resulting in more efficient data sharing.

● Integrated RNA assays, enhancing AI/ML module performance efficacy by at least 10%.

● Enabled AI/ML team in patient stratification with robust data pipelines for static and time series data points.

● Coordinated data governance and integration between AI/ML, data management, and DevOps functions, to bring clinical ML from prototype to production.

● Published author on ARSA gene research for metachromatic leukodystrophy, adding to rare disease drug development.

● Streamlined data analysis and visualization using R, Python, and MATLAB, improving data analysis speed by 70%.

● Developed workflows for pathogenic variant identification across 60 rare disorders, resulting in more accurate prediction of disease prevalence and incidence.

● Prototyped relational databases with Hail for efficient genomic data storage, improving data management and accessibility.

● Identified over forty putative conserved vertebrate phosphodiesterase proteins, using multiple sequence alignment and genomic evidence.

Topics: Introduction to RNA-Seq, Standard Workflows with DESeq2, Reference-Based Assembly, Exploratory Data Analysis, Statistical Analysis, Data Visualization, Gene Set Enrichment, Variant Annotation

● Presented publication-quality RNA-Seq data summaries and representations to post-doctoral associates

● Presented now-published preprint in contrast with established research at internal Journal Club to inform future experimental design in epigenetic editing.

General Biology II facilitator for two concurrent PLTL courses