By John Peach | August 28, 2018
Talk 1: Classification and Statistical Analysis of Cancer Mutations Scores By: Yemi Odeyemi (Ph.D. Candidate in Data Science at Chapman University)
The talk will describe part of Yemi’s doctoral work on building a statistical and predictive model to classify driver-passenger mutations. A Logit model is used with 10-fold cross-validation. The data was preprocessed to impute missing values using the rule-of-thumb approach, removal of redundant features and feature scaling. Feature selection was determined using a stepwise approach based on AIC. The objective was to determine the optimal class boundary for the probability for discretization. The models were evaluated with Receiver Operator Characteristics - Area under the curve (ROC-AUC) which is based on sensitivity and specificity.
Talk 2: How Reactivity Works in Shiny Apps By: Tuck Nugn (Senior Data Scientist at Oracle)
Reactivity is not a common programming paradigm but is an important part of the Shiny framework. In this talk we will deep dive into this technology and discuss: * What reactivity means in the context of Shiny apps * How reactive objects work in a practical sense * How you can use it to make your apps more efficient
Agenda: 6:30-7:00 Welcome and Networking 7:00-8:00 * Introduction * Talk 1: Classification and Statistical Analysis of Cancer Mutations Scores * Talk 2: How Reactivity Works in Shiny Apps 8:00-8:30 Clean-up and networking