1  Introduction

2 Welcome!

This is a book for the Advanced Data Science for Bio/Public Health/medicine classes. Since data science isn’t super well defined, advanced data science is even less so. My opinion is that we needed the umbrella term data science because there was a lot in the processes of using and analyzing data that got ignored in traditional disciplines and training programs. Of course, what got ignored is different depeneding on which discipline or program. I’m mostly going to focus on concepts and implementation that were historically ignored in our (JHU Biostat) program, which is heavily focused on biostatistical inference, probability modeling, public health/bio/medical data analyses and ML. Loosely, there’s three main thrusts of the notes: technique, concepts and methods and meta data science. By technique I mean tools, like webscraping. By concepts and methods I mean things like variational Bayes. Meta data science refers to things like the study of data science itself, meta science studies and ethics. However, since the goal is to put things together, these concepts get jumbled up a bit.

Ultimately, if I were forced to come up with a definition of data science, I would give: “Data science is the practice of putting all of the aspects of the productive use of data together.” So, statistical theory, computational theory, statistical methods, computational methods, statistical practice, computational practice, ML, AI, deployment, communication, data ethics, … are all part of data science: provided they’re needed to make productive use of data. So, in a sense, everyone who studies any of those topics in isolation is a data scientist. However, in this course the goal is to try to put those things together; that is, to land somewhere in the middle of the simplex rather than on an edge or point.

2.1 Requirements

I’m going to assume that you have a lot of basic data science tools down already. If not, here’s some notes. For this book you’ll need: prior programming experience, calculus, linear algebra, unix, python, R, basic AI, basic ML and basic statistics.

2.2 Organization

This is a two quarter course. The first quater is devoted to tools and the second is deveoted to theory. So the book is divided in half that way.

2.3 Reading

Read these papers

1. Tukey (1962)
2. Donoho (2017)
3. Leek and Peng (2015)
4. Leek and Peng (2015)
5. Kass et al. (2016)
6. Hicks and Irizarry (2018)
7. Hardin et al. (2015)

Donoho, David. 2017. “50 Years of Data Science.” Journal of Computational and Graphical Statistics 26 (4): 745–66.

Hardin, Johanna, Roger Hoerl, Nicholas J Horton, Deborah Nolan, Ben Baumer, Olaf Hall-Holt, Paul Murrell, et al. 2015. “Data Science in Statistics Curricula: Preparing Students to ‘Think with Data’.” The American Statistician 69 (4): 343–53.

Hicks, Stephanie C, and Rafael A Irizarry. 2018. “A Guide to Teaching Data Science.” The American Statistician 72 (4): 382–91.

Kass, Robert E, Brian S Caffo, Marie Davidian, Xiao-Li Meng, Bin Yu, and Nancy Reid. 2016. “Ten Simple Rules for Effective Statistical Practice.” PLoS Computational Biology. Public Library of Science.

Leek, Jeffery T, and Roger D Peng. 2015. “What Is the Question?” Science 347 (6228): 1314–15.

Tukey, John W. 1962. “The Future of Data Analysis.” The Annals of Mathematical Statistics 33 (1): 1–67.