Practicing AI + Biology Research

We are dedicated to understanding biological processes through interpretable machine learning. Our colleagues at the University of Chicago have conducted rigorous biological experiments over a span of two years. Following this, I was lucky to work with Prof. Oded Regev on the subsequent, in-depth analysis of the results. These collective efforts will culminate in a research paper and other materials.

Professor Oded Regev is truly exceptional. I gained invaluable insights and knowledge under his mentorship. His approachability is also noteworthy, as he kindly arranged for my office to be situated next to his at Courant, allowing us to meet 2-3 times per week. The entire research journey has been an enjoyable and enriching experience.

Reflecting on my work over the past year, several key insights have emerged.

Embracing the Challenges in AI + Biology

Given the intricate nature of biological data, it is often characterized by noise and complex structures. This necessitates a considerable amount of work in data cleaning, processing, and rigorous validation, among other responsibilities. We need to ensure the highest level of accuracy and reliability in our analyses, which ultimately contributes to advancements in our understanding of biological systems.

Navigating Idea Selection and Experiment Design

Most people would benefit greatly by spending more time on idea selection. It’s valuable to keep the big picture in mind, then zero in on those ideas that can stand the test of time. The following stage is to design and carry out experiments to test the effectiveness of these ideas. The importance of rigorous and comprehensive experiment design cannot be overstated. Even brilliant ideas can be ruined by a messy experiment design.

Promoting Effective Documentation and Coding Practices

Maintaining well-structured progress reports, comprehensive experimental records, and an elegant code style is essential in our research. These practices help us track project development, identify potential issues, effectively allocate resources, minimize the risk of errors, and ensure reproducibility. Embracing these habits not only improves the quality and efficacy of a project but also fosters seamless collaboration among team members and guarantees the project’s long-term success.