Detecting cancer at early stages and monitoring disease progression during treatment are critical to improving the chance of cure and decreasing cancer morbidity and mortality. However, early detection of cancer remains challenging, as most traditional methods are costly and only able to detect sizable tumors. Thus, we aim to develop a cost-effective and minimally invasive computational approach for pan-cancer detection and tumor burden estimation using data from ultra-low coverage whole genome sequencing of plasma cell-free DNA (cfDNA). In this project, I will work on developing and implementing statistical inference algorithms that integrate different cfDNA measurements such as CNV and fragment length to distinguish tumor-derived from normal-derived cfDNA and estimate tumor fraction.

I am participating in SuperUROP because I want to gain research experience in computational genomics. The topic and aims of this project are aligned with my interests in cancer biology and copy number variation analysis. I am excited to learn skills needed for conducting computational research. I also hope to apply and expand upon knowledge I have learned in biostatistics and machine learning classes by working on real‐world applications.