Math & Stats Colloquium: Jaime Davila
Speaker: Jaime Davila, Assistant Professor, St. Olaf College
Title: Data Science in Cancer Genomics: Filtering Artifactual Mutations in Clinical Archival Samples
Abstract: Somatic mutations are changes in DNA that occur during the lifetime of a cell and can result in cancer. Tests to detect somatic mutations are commonplace in the clinical treatment of cancer patients. Most clinical tests use archival pathology slides, which are preserved by a process known as Formalin-Fixation Paraffin Embedding (FFPE). An unfortunate consequence of the FFPE process is the generation of C to T mutations which are not present in the original sample. By using machine learning techniques, we recently identified a unique mutational signature present in FFPE tissues. In this talk we will describe excerno, a novel method for filtering FFPE artifacts. Excerno leverages mutational signatures, linear matrix decompositions, and Bayes’ theorem. We will describe each of these techniques and illustrate how to use them to solve this problem. Finally, we will show the strategy used for validating excerno and some of our findings.
Bio: Jaime Davila is an Assistant Professor at the Mathematics, Statistics and Computer Science Department at St. Olaf College. Jaime did his undergraduate studies in Math and Computer Science at the Universidad de los Andes in Bogotá, Colombia. He received his PhD in Computer Science from the University of Connecticut at Storrs and worked as an informatics specialist at Mayo Clinic, Rochester for nine years. Jaime’s primary research interest is in the creation and application of data science methodologies that allow the interpretation of genomic sequencing, particularly in the context of cancer genomics. In his spare time Jaime enjoys playing card games like old maid and crazy eights with his wife and two young daughters.