My work is at the intersection of biology, neuroscience, mathematics, and computer science. Although research articles have a clearly stated objective — identifying and resolving gaps in understanding or knowledge within a topic — not all research is explicitly applied. Most research falls into the category of basic science, which is the pure pursuit of knowledge in the hope that the future might hold significant applications (drugs, tools, technologies, etc.). Whereas basic science typically suggests future applications, there is rarely a well defined path. Machine Learning/AI and computer technologies, in general, are methods that immediately clarify the applications, and this has been my motivation in taking an integrative or interdisciplinary approach. Research, in my view, should always address the “How” questions: How will it be used? How does the work change the way we think about the problem? How does it make an impact? Below, click on the tiles to review some of my work. Send requests for PDF copies through the contact form.
Overviews
Attempts to remove technical details
Dissertation, Book Chapters, and Preprints
This is a selection of other publications including my PhD dissertation on computational neuroscience and artificial intelligence. It focuses on the problem of controlling insect-borne illnesses through neuroscience and AI. It begins with a discussion on insect behaviors like attraction and repellency and describes an approach to predict these behaviors using AI/Machine Learning. Next, because the approach showed promise, this suggests methods to induce or control the behavior. Namely, by studying how the insect nervous system detects and responds to airborne chemicals (volatiles) , AI algorithms can learn the chemical signatures that evoke specific behaviors (like insect repellency). It is then possible to search nature’s possibilities for similar but novel chemicals. We may be interested in doing so for a variety of reasons. Factors such as the human and/or ecological toxicity of a chemical as well as the cost of synthesizing it drive interest in better, safer, and more efficient chemicals. But this task is not easy. How does a chemist analyze millions of chemicals? And what would they look for? This is where AI enters.
The trained AI algorithm searched millions of chemicals for designer insect (or far broader, arthropod) repellents. After selecting promising candidates, a different AI algorithm was developed to simulate long-term toxicity on humans and the broader ecosystem. Collectively, the process involved searching tens of millions of chemicals, including nature’s pre-existing solutions such as hundreds of thousands of chemicals on human skin. The work has since been adapted for commercialization alongside partnering Fortune 500, 100, and 50 companies.
Interact with the titles to read
