Chuan Li Email & Phone Number

Redmond, Washington, Us

Science tech lead in AI Architecture & Strategy, Microsoft Research;Develop NLP solutions to provide more relevant search results for Bing users.

Stanford, Ca, Us

Develop cutting-edge technologies to study tumor progression and therapeutic responses using machine learning and statistical methods.1. Developed a Python pipeline including data cleaning, anomaly detection, statistics generation, and feature extraction to quantify genotype-specific therapeutic responses, reducing error rate of tumor call to 1/300 of existing methods. A VC-backed company (D2G Oncology) was founded based on the technology.2. Implemented a platform that integrates TCGA, GENIE and GDSC databases for data fetch, query, processing, and version control; generated the first pharmacogenomics map in lung cancer with high-throughput approaches, achieving >10x scale of existing work; validated predictions with clinical data.3. Implemented tools to automatically parse and analyze terabytes of DNA sequencing data to measure novel features for tumor suppressor effects; used unsupervised learning methods (PCA, k-means, UMAP, t-SNE) to reveal gene clusters in tumor initiation, progression, and metastasis.

Ann Arbor, Michigan, Us

Developed multiple analytical pipelines and machine learning methods in quantifying intergenic and intragenic interactions at a large-scale.1. Designed an information retrieval system to quantify fitness of >65,000 gene variants, achieving over 100x larger scale than any prior work. Discovered novel features and built ML models based on decision trees to predict the fitness of mutations, revealing deep insights on the underlying determinant of fitness landscapes, which revolutionized people’s understanding. First-author paper published in Science.2. Developed novel methods based on fractional factorial design and bootstrap to maximize the statistical power in detecting genetic interactions. Optimized the identification strategy with low FWER using various sample sizes and statistical models. Used rigorous testing to confirm its effectiveness.

Cn

1. Modeled drug metabolism in patients with liver diseases using multiple machine learning methods.2. Analyzed terabytes of large-scale omics data and identified multiple miRNAs-mRNA interactions.

Phd, Ecology And Evolutionary Biology

Master'S Degree, Statistics

Bachelor'S Degree, Biological Sciences