Similar color analysis based on deep learning (SCAD) for multiplex digital PCR via a single fluorescent channel
Similar color analysis based on deep learning (SCAD) for multiplex digital PCR via a single fluorescent channel
复制标题
DOI：
10.1039/d2lc00637e
复制
发表时间：
2022-09-02
期刊：
影响因子：
6.1
通讯作者：
Xu, Feng
中科院分区：
文献类型：
--
作者：
Cao, Chaoyu;You, Minli;Xu, Feng
数字PCR （dPCR）由于其高灵敏度和准确性，近年来引起了人们的广泛关注。然而，现有的dPCR依靠多色荧光染料和多个荧光通道来实现多重检测，导致检测成本增加，检测吞吐量有限。在这里，我们开发了一种基于深度学习的相似颜色分析方法，即SCAD，以在单个荧光通道中实现多重dPCR。作为演示，我们设计了一个基于微孔芯片的双工dPCR系统，用两种绿色荧光探针检测bla（NDM）和bla（VIM）两个基因，这两种荧光探针的发射颜色难以用传统的基于荧光强度的方法区分。为了验证深度学习算法区分相似颜色的可能性，我们首先应用t分布随机邻居嵌入（tSNE）对具有相似荧光的微孔进行聚类图。然后，我们在10,000个具有两种相似颜色的微孔上训练了视觉变压器（Vision Transformer, ViT）模型，并在262 - 202微孔上进行了测试。最后，训练后的模型被证明具有高度准确的分类能力（训练集和测试集均为>98%）和bla（NDM）和bla(VIM)（比率差）的精确量化能力
Digital PCR (dPCR) has recently attracted great interest due to its high sensitivity and accuracy. However, the existing dPCR depends on multicolor fluorescent dyes and multiple fluorescent channels to achieve multiplex detection, resulting in increased detection cost and limited detection throughput. Here, we developed a deep learning-based similar color analysis method, namely SCAD, to achieve multiplex dPCR in a single fluorescent channel. As a demonstration, we designed a microwell chip-based diplex dPCR system for detecting two genes (bla(NDM) and bla(VIM)) with two kinds of green fluorescent probes, whose emission colors are difficult to discriminate by traditional fluorescence intensity-based methods. To verify the possibility of deep learning algorithms to distinguish the similar colors, we first applied t-distributed stochastic neighbor embedding (tSNE) to make a clustering map for the microwells with similar fluorescence. Then, we trained a Vision Transformer (ViT) model on 10 000 microwells with two similar colors and tested it with 262 202 microwells. Lastly, the trained model was proven to have highly accurate classification ability (>98% for both the training set and the test set) and precise quantification ability on both bla(NDM) and bla(VIM) (ratio difference