Related Experiment Video
Updated: Jul 16, 2026
06:57
Rapid, Scalable Assembly and Loading of Bioactive Proteins and Immunostimulants into Diverse Synthetic Nanocarriers Via Flash Nanoprecipitation
Published on: August 11, 2018
Self-driving lab for the data-driven design of single-chain polymer nanoparticles
Alexander Suponya1,2, Elena Di Mare1, Cesar Ramírez1
1Department of Biomedical Engineering, Rutgers, The State University of New Jersey, Piscataway, NJ 08854, USA.
Cell Press Blue
|July 15, 2026
Summary
Scientists developed an automated workflow using active learning to discover structure-property relationships in single-chain polymer nanoparticles (SCNPs). This platform accelerates the design of SCNPs for biomedical applications by predicting and synthesizing tailored polymer structures.
Area of Science:
- Polymer Chemistry and Materials Science
- Nanotechnology
- Biomedical Engineering
Background:
- Single-chain polymer nanoparticles (SCNPs) offer tunable properties and structural complexity for protein mimicry.
- Discovering structure-property relationships in SCNPs is crucial for tailored applications.
- Current methods for SCNP design are often slow and labor-intensive.
Purpose of the Study:
- To develop a high-throughput, autonomous workflow for discovering SCNP structure-property relationships.
- To iteratively predict and synthesize SCNPs using active learning.
- To create a platform for designing SCNPs with specific properties for biomedical use.
Main Methods:
- An automated system integrating a liquid-handling robot, custom photo-polymerization lightbox, dynamic light scattering (DLS) reader, and robotic arm was developed.
- Photoinduced electron/energy transfer reverse addition-fragmentation chain transfer (PET-RAFT) polymerization was employed.
- Gaussian process regressor (GPR) models trained on seed libraries were optimized using Bayesian optimization (BO).
Main Results:
- The active learning workflow successfully identified structure-property relationships for SCNPs.
- Iterative synthesis and model refinement improved prediction accuracy.
- The impact of monomer content on particle size (Rh) was effectively modeled.
Conclusions:
- An automated polymer discovery platform was successfully established for SCNP synthesis.
- This platform enables efficient design of SCNPs with tailored structures.
- The workflow serves as a prototype for advancing SCNP design in biomedical applications.
