Mackenzie Noon

Ph.D. candidate in the Reilly Lab, Department of Genetics, Yale University.

Publications

Bypassing current biosecurity screens with AI-designed proteins, and closing the gap with edge-AI functional screening. Noon M, Oualline G. Apart Research AIxBio Hackathon, 2026. [writeup]
Global Evaluation of Congenital Heart Disease-Associated Non-Coding Variants. Peña-Martínez EG, Sharma S, Medina-Feliciano JG, et al., Noon M, et al., Bhimsaria D, Rodríguez-Martínez JA. bioRxiv, 2025. [doi] [preprint]
Identifying non-coding variant effects at scale via machine learning models of cis-regulatory reporter assays. Butts JC, Rong S, Gosai SJ, Castro RI, Noon M, Adeniran K, Ghosh R, Sabeti PC, Tewhey R, Reilly SK. bioRxiv, 2025. [doi] [preprint]
Taming the transposon: H3K9me3 turns foe to friend in human development. Chitrakar A, Noon M, Xiao AZ. Cell Stem Cell, 2022. [doi] [published]
A mouse-specific retrotransposon drives a conserved Cdk2ap1 isoform essential for development. Modzelewski AJ, Shao W, Chen J, et al., Noon M, et al., Risso D, He L. Cell, 2021. [doi] [published] [preprint]

Common Mechanism in a Box: point-of-synthesis DNA screening. Noon M, Alexanian T. CBAI AIxBiosecurity Research Fellowship, 2026. [poster]
Capiti: Improving Biosecurity Screens With Edge AI. Noon M, Oualline G. CyberBio 2026: First International Workshop on Cyberbiosecurity, co-located with IEEE S&P, 2026. [poster]
Development of single-cell massively parallel reporter assays for measurement of disease-associated variant effects. Noon M, Gilbertson EN, Reilly SK. Boehringer Emerging Scientist Talent (BEST) Program, 2025. [poster]

commec-gui — Offline kiosk and LAN web interface, merged into IBBIS's Common Mechanism.
commec-in-a-box — Pre-imaged mini-PC appliance that runs the Common Mechanism on site, built and distributed by a reproducible image-minting pipeline.
scMPRAforge — Python package for simulating, designing, and testing single-cell massively parallel reporter assays. Manuscript in preparation; release pending.
tensorzinb-plusplus — Fork of TensorZINB adding current TensorFlow support, sparse input matrices, and sample-weighted zero-inflated negative binomial regression.

MPRA power analysis (pdf) — lecture notes on error control, p-values, and simulation-based power for reporter assays. August 2026.

Last updated August 2026.