Odor Divergence of Enantiomers & Pyrfume

My curiosity about how our inner experiences and the external environment interact and shape one another drove me to pursue computational neuroscience and perception research. I did my honors thesis on using machine learning to predict the odor divergence of enantiomers based on the molecules' structural properties.

My Role: Honors Senior Thesis

Status: Completed Report, Presented Poster, Published interactive tutorial (seen below), Published paper in BioRxiv, Github

Interactive Tutorial with Jupyter Book

Poster

OdorDivergenceOfEnantiomersPoster.pdf

Presented Poster At:

Osmonauts Meeting, Cold Spring Harbor, New York, November 2021. Student, S. “Predicting Odor Divergence of Enantiomers'' (poster).
School of Life Sciences Undergraduate Research, virtual, April 2021. Student, S. “Low-Dimensional Embeddings of Mouse Olfactory Perceptual Space” (poster).
Osmonauts Meeting, virtual, August 2020. Student, S. “Predicting Odor Divergence of Enantiomers'' (poster).
International Symposium on Olfaction and Taste, virtual, August 2020. Student, S. “Predicting Odor Divergence of Enantiomers” (poster).

The data collected and code written in this project lent itself to the contributions to Pyrfume, a framework of olfactory datasets and modules.

Publication: J.B. Castro*, T.J. Gould*, R. Pellegrino, Z. Liang, L. Coleman, F. Patel, D. Wallace, T. Bhatnagar, J.D. Mainland, R.C. Gerkin, “Pyrfume: A Window to the World’s Olfactory Data”, published [p], 2022.

Future/Challenges/Learnings: This project gave me valuable insight into the predictive power of machine learning algorithms, along with challenges like ensuring a high-quality, complete dataset. I want to continue exploring perception by combining perceptual research with interactive tools that help visualize and reflect on the connection between our inner and outer worlds. I hope to gain more experience in integrating signals, sensors, and data into hardware systems to create well-informed perceptual experiences.