Bio

I'm a PhD student in the Institute for Computational and Mathematical Engineering (ICME) at Stanford University, mentored by Prof. Mignot. My research is at the intersection of artificial intelligence and sleep medicine, focusing on developing predictive models for circadian rhythms and sleep debt from proteomics data. I adopt a problem-oriented approach, selecting methods based on the data and research questions at hand. My techniques range from linear regression to sophisticated deep learning frameworks, aiming to extract maximal insights from the data. I also explore the use of unsupervised and semi-supervised learning, and am interested in the applications of multimodal and foundation models in biology.

Contact

  • Academic
    aspecht@stanford.edu
    University - Student Department: ICME Operations Position: Graduate

Additional Info

All Publications

  • CIRCADIAN PROTEINS AND CIRCADIAN PHASE ESTIMATION IN HEALTHY ADULTS Specht, A., Saha, P., Cederberg, K., Bueno, F., Arrona-Palacios, A., Pardilla-Delgado, E., Ruiz-Herrera, N., Zitting, K., Kramer, A., Zeitzer, J., Czeisler, C., Duffy, J., Mignot, E. OXFORD UNIV PRESS INC. 2024

    View details for DOI 10.1093/sleep/zsae067.0009

    View details for Web of Science ID 001262172000010

  • Circadian protein expression patterns in healthy young adults. Sleep health Specht, A., Kolosov, G., Cederberg, K. L., Bueno, F., Arrona-Palacios, A., Pardilla-Delgado, E., Ruiz-Herrera, N., Zitting, K. M., Kramer, A., Zeitzer, J. M., Czeisler, C. A., Duffy, J. F., Mignot, E. 2023

    Abstract

    To explore how the blood plasma proteome fluctuates across the 24-hour day and identify a subset of proteins that show endogenous circadian rhythmicity.Plasma samples from 17 healthy adults were collected hourly under controlled conditions designed to unmask endogenous circadian rhythmicity; in a subset of 8 participants, we also collected samples across a day on a typical sleep-wake schedule. A total of 6916 proteins were analyzed from each sample using the SomaScan aptamer-based multiplexed platform. We used differential rhythmicity analysis based on a cosinor model with mixed effects to identify a subset of proteins that showed circadian rhythmicity in their abundance.One thousand and sixty-three (15%) proteins exhibited significant daily rhythmicity. Of those, 431 (6.2%) proteins displayed consistent endogenous circadian rhythms on both a sleep-wake schedule and under controlled conditions: it included both known and novel proteins. When models were fitted with two harmonics, an additional 259 (3.7%) proteins exhibited significant endogenous circadian rhythmicity, indicating that some rhythmic proteins cannot be solely captured by a simple sinusoidal model. Overall, we found that the largest number of proteins had their peak levels in the late afternoon/evening, with another smaller group peaking in the early morning.This study reveals that hundreds of plasma proteins exhibit endogenous circadian rhythmicity in humans. Future analyses will likely reveal novel physiological pathways regulated by circadian clocks and pave the way for improved diagnosis and treatment for patients with circadian disorders and other pathologies. It will also advance efforts to include knowledge about time-of-day, thereby incorporating circadian medicine into personalized medicine.

    View details for DOI 10.1016/j.sleh.2023.10.005

    View details for PubMedID 38087675

https://orcid.org/0000-0001-7301-0419