What we're about
Upcoming events (5)
Researchers utilize high-dimensional biological data for questions ranging from identifying groups of interest to extracting biologically meaningful representations. This hands-on unsupervised machine learning workshop will introduce concepts such as cluster validation and stability, dimensionality reduction, and representation learning for transcriptomics. About our speaker: Jaclyn Taroni is a Principal Data Scientist at the Childhood Cancer Data Lab, an initiative of Alex's Lemonade Stand Foundation. She was a Postdoctoral Researcher in the Greene Lab at the University of Pennsylvania Perelman School of Medicine, where she worked on cross-platform normalization of gene expression data and unsupervised transfer learning for rare diseases. She conducted her thesis work in the Whitfield Lab at the Geisel School of Medicine at Dartmouth studying the rare autoimmune disease systemic sclerosis, where she developed novel frameworks for analyzing high-throughput molecular data from multiple tissues, clinical manifestations, and drug trials. (http://www.jaclyn-taroni.com/)
DESCRIPTION: Decision trees are easy to understand and form the basis for random forest, a powerful machine learning technique. We will build decision tree and random forest models to discover trends using the Philadelphia Police Department’s Complaints Against Police dataset. Please note that we are using the dataset to illustrate statistical and programming techniques and do not have the background in criminal justice to draw any conclusions from our models. SPEAKER: Alexandra Parfitt is a graduate student in applied statistics at Villanova University. She is completing a co-op at GlaxoSmithKline in chemical, manufacturing and controls statistics in the biopharmaceutical division. She entered data science and statistics after working in the humanities and she holds an AB from the University of Chicago and a PhD from Yale in comparative literature. She lives in Ardmore with her husband and their four-year-old daughter.
Introduction This workshop is for people looking to learn how to make their own R packages and learn how to use “usethis” and “devtools” for package development. The workshop will cover handy one time functions (i.e., use_this::create_package) as well as functions used continuously throughout package development (i.e., devtools::document). At the end of the hour you should have a working, well-documented package with a single function, as well as a better understanding of the files and file structure required for R packages. This workshop is suitable for beginner to intermediate R users. Attendees should be familiar with functions, but will not be writing their own function in this workshop. Familiarity with pipe or tidyverse is helpful. Speaker Shannon Pileggi is an enthusiastic statistical collaborator and professional educator with over ten years of experience partnering on data analysis with diverse stakeholders in corporate settings, tech, public health, and clinical research. Shannon’s previous work experiences include supporting research on parasitic diseases at the CDC, developing a pilot college-wide intro stats course at the Institute for Quantitative Theory and Methods at Emory University, and instructing R/SAS programming, intro stats, and survival analysis courses at California Polytechnic State University, San Luis Obispo. Shannon is currently a Statistician at Adelphi Research in Doylestown, PA working in pharmaceutical market research. Shannon’s day-to-day in R include: fine-tuning ggplots to convey analysis results, contributing to an internal package to automate routine analyses/tasks, streamlining work-flows for repetitive analyses, and developing shiny apps.