Instructor: Sheri Sanders
Office: Galvin 252 (behind 2nd fl aud)
Office hours: By Request
Email: [email protected]
Lecture: 11a-1p, MWF
11/4 - 11/22 (three weeks)
204 Jordan Hall
This module focuses on providing researchers with the skills and knowledge necessary to critically read, interpret, and evaluate machine learning applications in biology. Machine learning has become a common part of many analyses, from gene annotation to drug discovery. Understanding and evaluating these applications requires an understanding of machine learning basics, limitations, and biases. The emphasis is on enhancing analytical thinking and understanding the potential challenges and limitations of applying ML techniques in biological research (this course is concept based and will not include coding). In this three week module, we will cover common machine learning methods and how they are evaluated, their application to various aspects of biology, and learn to critique research based on these techniques.
For various topics, materials will be posted in the form of videos, overviews, etc. to set the stage for talking about papers each meeting.
In class we will discuss topics, do activities to build understanding, and critique papers. In the first week, we will get up to speed on machine learning, then build toward a class-drafted critique guide in week two. We will then evaluate current topics and research papers in the third week. The goal of these discussions will be to critique the portions that we have covered (data, bias, methods, etc), building up a repertoire of common points to consider while reading these papers. Students will then select papers (with the help of the instructor if needed) based on a topic that is of interest to them.
The final assessment of these skills in understanding and evaluating machine learning papers in biology will be a final paper critique on a paper selected from a list provided or one selected and approved by the instructor. Students will present topics and submit a short report for a final grade.
After this module, students should be able to:
Recommended Preparation: An understanding of generic data and basic statistic is required. This is NOT a coding class, so technical skills are not required!