Center for Complex and Active Materials MRSEC的动态

Multi-Objective Design of DNA-Stabilized Nanoclusters findings published in ACS Publications. DNA-stabilized silver #nanoclusters (AgN-DNAs) exhibit sequence-dependent compositions and fluorescence colors. Recently, high-throughput experiments combined with supervised machine learning have enabled the design of #DNA templates that tune AgN-DNA properties, including near-infrared (NIR) emission, which shows promise for deep tissue bioimaging. Led by?Prof. Stacy Copp,?the IRG2 team and collaborators developed a multiobjective, continuous-property design model for AgN-DNAs using variational autoencoders (VAEs) with automatic feature extraction. This generative model learns both the forward mapping from DNA sequence to AgN-DNA properties and the inverse mapping from properties to sequence. It was trained on experimental data linking DNA sequences to AgN-DNA fluorescence characteristics. Experimental validation demonstrated the model’s ability to design AgN-DNAs with enhanced emission, including bright NIR-emitting AgN-DNAs, which were produced with four times greater abundance than those in the training set. This generative approach can be extended to a variety of biomolecular systems with sequence-dependent properties, facilitating precise design of emerging nanomaterials. The theoretical work was led by MRSEC junior research fellow?Peter Mastracco, who recently earned his Ph.D., in collaboration with?Petko Bogdanov’s experimental group at the University at Albany-SUNY.

  • 该图片无替代文字

要查看或添加评论,请登录