Abstract. Genome and tumor sequencing is yielding hundreds of variants associated with disease across dozens of genes, but determining causality and the functional impact of each variant is a major bottleneck. We propose morphological profiling as a rapid and inexpensive method to systematically map the impact of variants. Morphological profiling extracts single-cell measurements from microscopy images to compute signatures of treatments at high-throughput. Such signatures encode variations in cell state that can be analyzed to identify unexpected correlations between chemical and genetic perturbations. We are developing computational tools, including deep learning-based methods, to discern the functional impact of variants of unknown significance in lung cancer, and to explore how morphological profiles can complement gene expression profiles to this end.
Short Bio. Juan Caicedo is a postdoctoral researcher at the Broad Institute of MIT and Harvard, where he investigates the use of deep learning to analyze microscopy images. Previous to this, he studied object detection problems in large scale image collections also using deep learning, at the University of Illinois in Urbana-Champaign. Juan obtained a PhD from the National University of Colombia and completed research internships in Google Research, Microsoft Research, and Queen Mary University of London as a grad student, working in problems related to large scale image classification, image enhancement, and medical image analysis. His research interest include computer vision, machine learning and computational biology.