Computational Design of Synthetic Optical Barcodes in Microdroplets

• Published in: Advanced optical materials 2024-04, Vol.12 (12), p.n/a
• Main Authors: Kawasaki, Fumiko, Mimori, Takahiro, Mori, Yuka, Aburatani, Hiroyuki, Yachie, Nozomu, Sato, Issei, Ota, Sadao
• Format: Article
• Language: English
• Subjects: Bar codes; Combinatorial analysis; computational material design; Decoding; droplet microfluidics; Gene sequencing; machine learning; Microparticles; multimodal barcoded microparticles; Neural networks; optical barcoding
• ISSN: 2195-1071; 2195-1071
• DOI: 10.1002/adom.202302564

Barcodes are useful for identifying objects across time, space, and information modalities. However, materializing and decoding optical and multimodal barcodes on microscopic objects remains difficult despite the increasing need for multiplexed cell analysis. Here, a computational design of randomly combinatorial is presented, yet decodable barcodes in microdroplets. The design is based on a novel Real2Sim2Real framework: it first collects experimental images of optically distinct microparticles, then simulates massive combinatorial images by randomly assembling the imaged particles to train a neural network‐based decoder. It is demonstrated that the decoder, even though trained via simulation, accurately identifies the randomly assembled particles in real hydrogel microdroplets. It also shows that the microdroplets with an additional DNA barcoding functionality are applicable to individually link independently measured microscopic images and transcriptome profiles of pooled single cells. The first computational design, based on Real2Sim2Real machine learning is presented, for random synthetic optical barcoding of microdroplets and demonstrates their real‐world implementation. As an application of the computationally designed droplets, it also shows their usage for tracking individual suspended cells across different instruments and information modalities (optical microscopy and genome sequencing).