Matt Thomson is an American computational biologist, academic, and entrepreneur. He works in the fields of computational biology, biophysics, and machine learning.[1]
Matt Thomson's academic journey began at Harvard University, where he pursued his undergraduate education and graduated magna cum laude with an AB in Physics in 2001.[2]
Thomson continued at Harvard University for graduate studies in Biophysics, completing his Ph.D. in 2011. His doctoral research focused on the mathematical modeling and analysis of biochemical networks, aiming to understand the control mechanisms[3] behind cellular decision-making processes.[4][5]
After earning his Ph.D., Thomson received an independent fellowship at the University of California, San Francisco (UCSF), where he worked on developing mathematical methods to model cell fate determination and tissue self-organization.[6][7]
Research and career
Matt Thomson is a faculty member in the field of computational biology at the California Institute of Technology (Caltech).[8] He is also the principal investigator of SPEC, the Beckman Center for Single Cell Profiling and Engineering at Caltech, and an investigator with the Heritage Medical Research Institute.[9] During his tenure at Caltech, Thomson mentored several PhD students, including Guruprasad Raghavan, who later co-founded Yurts with him in 2022. His research group studies "living algorithms," focusing on the information processing strategies used by cells and organisms to interact with their environment.[10] This interdisciplinary area includes computational biology, physics, machine learning, and neuroscience.[11]
His research in bio-inspired machine learning algorithms, such as FIP, Herd, and spatial predictive coding, has contributed to advancements in neural network flexibility.[12] In addition, he developed mathematical models for predicting cellular responses to external stimuli, including D-SPIN, Popalign, and ActiveSVM.[13]
Additionally, Thomson's research includes the creation of optically programmable active materials by engineering light-switchable proteins integrated into cytoskeletal networks.[14]
Awards
Matt Thomson has received several awards and honors for his work in computational biology, machine learning, and systems biology. In 2019, he received the Packard Fellowship.[15] He also received the NIH Early Independence Award in 2011.[16] Additionally, Thomson received the NIH Transformative R01 in 2023 and the Okawa Research Award in 2020.
Raghavan, Guruprasad; Thomson, Matt (2019). "Neural networks grown and self-organized by noise". arXiv:1906.01039 [cs.NE].
Raghavan, Guruprasad; Tharwat, Bahey; Hari, Surya Narayanan; Satani, Dhruvil; Thomson, Matt (2022). "Engineering flexible machine learning systems by traversing functionally-invariant paths". arXiv:2205.00334 [cs.LG].
Jiang, Jialong; Chen, Sisi; Tsou, Tiffany; McGinnis, Christopher S.; Khazaei, Tahmineh; Zhu, Qin; Park, Jong H.; Strazhnik, Inna-Marie; Hanna, John; Chow, Eric D.; Sivak, David A.; Gartner, Zev J.; Thomson, Matt (2023-05-20). "D-SPIN constructs gene regulatory network models from multiplexed scRNA-seq data revealing organizing principles of cellular perturbation response". bioRxiv10.1101/2023.04.19.537364.
Ross, Tyler D., Heun Jin Lee, Zijie Qu, Rachel A. Banks, Rob Phillips, and Matt Thomson. 2019. “Controlling Organization and Forces in Active Matter Through Optically-Defined Boundaries”. Nature 572 (7768): 224–29. doi.org/10.1038/s41586-019-1447-1
Raghavan, G., Tharwat, B., Hari, S.N. et al. Engineering flexible machine learning systems by traversing functionally invariant paths. Nat Mach Intell 6, 1179–1196 (2024). doi.org/10.1038/s42256-024-00902-x
Gornet, J., Thomson, M. Automated construction of cognitive maps with visual predictive coding. Nat Mach Intell 6, 820–833 (2024). doi.org/10.1038/s42256-024-00863-1
Yang, F., Liu, S., Lee, H.J. et al. Dynamic flow control through an active matter programming language. Nat. Mater. (2025). doi.org/10.1038/s41563-024-02090-w
↑ Raghavan, Guruprasad; Thomson, Matt (2019-06-03). "Neural networks grown and self-organized by noise". arXiv:1906.01039 [cs.NE].
↑ Gornet, James A.; Thomson, Matt (2024-04-18). "Automated construction of cognitive maps with visual predictive coding". bioRxiv10.1101/2023.09.18.558369.
↑ Jiang, Jialong; Chen, Sisi; Tsou, Tiffany; McGinnis, Christopher S.; Khazaei, Tahmineh; Zhu, Qin; Park, Jong H.; Strazhnik, Inna-Marie; Hanna, John; Chow, Eric D.; Sivak, David A.; Gartner, Zev J.; Thomson, Matt (2023-05-20). "D-SPIN constructs gene regulatory network models from multiplexed scRNA-seq data revealing organizing principles of cellular perturbation response". bioRxiv10.1101/2023.04.19.537364.
This page is based on this Wikipedia article Text is available under the CC BY-SA 4.0 license; additional terms may apply. Images, videos and audio are available under their respective licenses.
