Despite the public health emphasis on population-level cancer screening in recent decades, adherence remains lower than desired, and cancer is often detected too late for successful treatment. In this talk, we will present Freenome’s multiomics platform that detects key biological signals from blood. The platform integrates assays for cell-free DNA, methylation, and proteins with advanced computational biology and machine learning techniques to understand additive signatures for early cancer detection. Furthermore, we will present biological insights revealed by latent representations from factor analysis and convolutional neural networks of genomic and proteomic data. These methods can be applied to other cancer types to learn biologically meaningful latent representations and to shed light on immune processes involved in cancer biology.

Tzu-Yu (Joyce) Liu is currently a staff machine learning scientist and senior manager at Freenome. Prior to her current position, she was a Simons Postdoctoral Fellow in the Departments of Mathematics and Biology at the University of Pennsylvania and with the Electrical Engineering and Computer Sciences (EECS) Department at the University of California, Berkeley, working with Professor Yun S. Song. Tzu-Yu received her B.S. from the National Taiwan University (2007) and her Ph.D. from the University of Michigan, Ann Arbor (2013), both in Electrical Engineering. Her Ph.D. research was in the area of statistical signal processing, under the supervision of her advisor Professor Alfred O. Hero III, and was co-advised by Professor Clayton D. Scott. She is a researcher in machine learning and computational biology. Her research includes statistical learning from high dimensional and small sample size problems, optimization, structured variable selection, density estimation, functional data analysis, transfer learning, multi-task learning, data fusion, convolutional neural network and their applications to biomedical data. These research experiences involved analysis of large datasets, e.g., genomic sequencing data, 3D/4D microscopy images, 3D human faces, and time series of electrocardiograms.