UQ-bio Summer School

2.1 – Basics of Machine Learning (William Raymond)

Lecture 2.1

Title: Learn basic machine learning tools for supervised and unsupervised analyses of single-cell data

Lecturer: Will Raymond

Lecturer Website: https://www.engr.colostate.edu/~munsky/

Lecturer Email: wsraymon@rams.colostate.edu

Learning Objectives:

      • Learn the basic types of machine learning (Unsupervised, Supervised, Decision) and learn when each one should be applied.

      • Test out simple classifiers in Tensorflow on a standard dataset (Iris and oxford flowers 17)

      • Extend these classifiers and try them on a simulated nascent chain tracking dataset. 

      • Visualize the decision boundaries of a FFNN on some 2D data.

Will Raymond graduated in May 2017 from Colorado State University with a Bachelors in Biomedical Engineering and a Bachelors in Chemical and Biological Engineering.

Will has been a PhD student in Dr. Munsky’s lab since 2019 working on a myriad of projects. He is currently the lead developer on the rSNAPsim, an open source Python TASEP mRNA simulation package. In addition to that, he also works on implementing machine learning for nascent chain tracking fluorescent spot identification. Additionally, a recent area of focus has been modeling tRNA abundances and their effect on translation dynamics.

His primary area of interest is the intersection of machine learning, bioinformatics, and RNA biology. He is most excited about novel ncRNA discovery and advancements in RNA structural identification and disease classification. 

Title: Learn basic machine learning tools for supervised and unsupervised analyses

Abstract: This lecture is an introduction to machine learning starting from the two classes of supervised and unsupervised learning. For each, we will go over some of the introductory first pass algorithms such as PCA and simple perceptrons. Additionally, at the end of the lecture we will provide a list of common loss functions, optimizers, and activation functions for classification with Neural Networks. 

Suggested Reading or Key Publications:

Links to Relevant Software: 

    • e
    • f