Multi-Objective Design of DNA-Stabilized Nanoclusters Using Variational Autoencoders With Automatic Feature Extraction
Year: 2024
Type: article
Abstract: High Resolution Image Download MS PowerPoint Slide DNA-stabilized silver nanoclusters (Ag N -DNAs) have sequence-tuned compositions and fluorescence colors. High-throughput experiments together with supervised machine learning models have recently enabled design of DNA templates that select for Ag N -DNA properties, including near-infrared (NIR) emission that holds promise for deep tissue bioimaging. However, these existing models do not enable simultaneous selection of multiple Ag N -DNA properties, and require significant expert input for feature engineering and class definitions. This work presents a model for multiobjective, continuous-property design of Ag N -DNAs with automatic feature extraction, based on variational autoencoders (VAEs). This model is generative, i.e., it learns both the forward mapping from DNA sequence to Ag N -DNA properties and the inverse mapping from properties to sequence, and is trained on an experimental data set of DNA sequences paired with Ag N -DNA fluorescence properties. Experimental testing shows that the model enables effective design of Ag N -DNA emission, including bright NIR Ag N -DNAs with 4-fold greater abundance compared to training data. In addition, Shapley analysis is employed to discern learned nucleobase patterns that correspond to fluorescence color and brightness. This generative model can be adapted for a range of biomolecular systems with sequence-dependent properties, enabling precise design of emerging biomolecular nanomaterials.
Source: ACS Nano
Language: English
Subfield: Materials Chemistry
Field: Materials Science
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid: $4,500
Funders: National Science Foundation, Silicon Valley Community Foundation, “la Caixa” Foundation, Chan Zuckerberg Initiative, Fundación Bancaria Caixa d'Estalvis i Pensions de Barcelona, University of California, Irvine, Office of the President, University of California, Division of Materials Research, Division of Chemical, Bioengineering, Environmental, and Transport Systems
Awards: G1135980590, UCI MRSEC: Materials Discovery Through Atomic Level Structural Design and Charge Control, G3227015700, G3511273693, G4228951661, G5051084384, Collaborative Research: New class of near-infrared fluorophores derived from DNA-templated silver clusters for deep tissue imaging, UC Collaborative for AI-enabled Materials Exploration and Optimization (UC-CAMEO), G97908232
OpenAlex ID: W4402574012