Thesis Defense In Chemistry, Presented by Joseph Hammer
Machine learning and Bayesian modeling for quantitative biological imaging: from single molecule localization microscopy to mass spectrometry imaging
Presented By: Joseph L. Hammer
Abstract:
The aim of this work is to establish and demonstrate unbiased, reproducible, and robust approaches to analyzing large and complex data from biological images. With advances in imaging techniques, more comprehensive and detailed data is produced than ever before. Extracting meaning from this data requires careful consideration to avoid overfitting while also maximizing the information that can be gained from the data. Herein, we present two forms of complex and challenging biological imaging data and propose approaches that can be used to extract key information from this data. By leveraging machine learning and Bayesian modeling, we demonstrate robust analysis pipelines for single molecule localization microscopy (SMLM) and mass spectrometry imaging (MSI) data. Specifically, for SMLM we developed a clustering tool enabling the Bayesian optimized selection of density-based clustering algorithm parameters in an unbiased manner that maximizes the density-based cluster validation score. For MSI, we leveraged supervised learning and a Bayesian hierarchical model to identify the spatially dependent role of mutant p53 in modulating key phospholipids and metabolites in 3D multicellular breast cancer spheroids.