Olivia Martinez

• Developed machine learning pipeline for predicting immunotherapy response using multi-modal patient data, improving prediction accuracy by 35% over existing methods
• Led analysis of 500+ patient tumor samples using single-cell RNA-seq, identifying novel T-cell exhaustion signatures that informed 3 clinical trial designs
• Built automated quality control system for genomic data processing, reducing manual review time from 40 hours to 2 hours per study
• Collaborated with 8 cross-functional teams to integrate computational findings into drug development decisions, contributing to 2 IND applications
• Mentored 3 junior bioinformaticians and established best practices for reproducible research workflows

• Analyzed whole-genome sequencing data from 2,000+ rare disease patients using GATK pipeline, identifying causal variants in 45% of cases
• Developed R Shiny application for interactive variant interpretation, enabling clinical geneticists to review cases 60% faster
• Implemented parallelized variant calling workflow using Cromwell/WDL, reducing processing time from 72 hours to 8 hours per sample
• Contributed to Cancer Genome Atlas (TCGA) pan-cancer analysis, processing 10TB+ of multi-omics data across 33 cancer types
• Published 4 first-author papers on computational methods for rare variant interpretation

• Designed deep learning architecture for drug-target interaction prediction, achieving 0.92 AUC on independent test sets
• Integrated chemical structure data with protein sequence features using graph neural networks, improving binding affinity predictions by 28%
• Analyzed RNA-seq data from 300+ cancer cell lines treated with experimental compounds, identifying synergistic drug combinations
• Created automated pipeline for processing ChIP-seq data, standardizing analysis across 15+ collaborative projects
• Presented research findings at 6 international conferences, winning 2 best poster awards