In 1958, Francis Crick’s central dogma laid the foundation for modern biomedical science by revealing the intricate connections between DNA, RNA, and proteins. This revelation has captivated researchers around the world ever since. As biotechnologies advance, scientists gain the ability to scrutinize the sequences, structures, and abundance of these critical molecules. Today’s next-generation sequencing technologies provide unprecedented opportunities to delve deeply into disease mechanisms, but they also introduce a host of statistical challenges.

My research aims to overcome these statistical hurdles, particularly in the realm of oncology. Our mission is to develop accessible and interpretable machine learning tools that bridge the gap between statistical learning methods and pratical biomedical applications, ultimately democratizing precision medicine for all.

Research Projects

Ongoing Projects

  • Single cell RNASeq based cancer model selection
  • Single cell tumor malignancy map construction
  • Evaluation of normalization methods on miRNA sequencing data

Collaboration

  • Great Lakes Breast Cancer Consortium data analysis
  • Outcomes after sentinel lymph node biopsy and radiotherapy in older women with early-stage, estrogen receptor–positive breast cancer

Methodology

  • Multi-study multi-class concordance analysis
  • Congruence analysis between cell lines and human tissues
  • Constrained Gaussian mixture model

Awards