放大器
计算机科学
多路复用
计算生物学
解码方法
深度学习
多重聚合酶链反应
多路复用
多重连接依赖探针扩增
核酸
人工智能
底漆(化妆品)
聚合酶链反应
管道(软件)
编码(内存)
特征(语言学)
深度测序
集合(抽象数据类型)
模式识别(心理学)
纳米孔
聚合酶
分子生物学
生物
灵敏度(控制系统)
Cre重组酶
人工神经网络
作者
Md. Ahasan Ahamed,Muhammad Asad Ullah Khalid,Wahid Uz Zaman,Anthony J. Politza,Weihua Guan
摘要
Single-reaction multiplex nucleic acid testing (NATs) must both co-amplify multiple targets and distinguish the resulting amplicons within the product mixture. Most conventional assays achieve this distinction using target-specific reporters or other external encoding elements. Here, we introduce a primer-programmed multiplexing strategy in which primer design generates target-associated amplicons with predefined lengths that are read electrically with nanopores. In a proof-of-concept implementation, one-pot recombinase polymerase amplification (RPA) generated four amplicons with predefined product lengths from Monkeypox (Mpox), Cowpox, Vaccinia, and human RNase P. The resulting amplicons were analyzed with glass nanopores, and individual translocation events were classified by a deep neural network (DNN) based on a six-dimensional feature set derived from nanopore recordings. Event-level decoding achieved >95% accuracy, >94.7% sensitivity, and >98.3% specificity, and aggregation of classified events yielded sample-level calls with >93.3% sensitivity and >93.9% specificity in blinded samples, with qualitative calls matching polymerase chain reaction (PCR) results across the 9 PCR-benchmarked samples. These results demonstrate a probe-free multiplexing architecture in which target identity is programmed through primer-defined amplicon design and decoded from the resulting multidimensional nanopore response using a DNN, eliminating the need for target-specific reporter chemistry and opening new opportunities for all-electrical multiplex nucleic acid testing based on unmodified amplification products.
科研通智能强力驱动
Strongly Powered by AbleSci AI