Related Experiment Video
Updated: Jan 12, 2026
15:10
From Fast Fluorescence Imaging to Molecular Diffusion Law on Live Cell Membranes in a Commercial Microscope
Published on: October 9, 2014
11.9K
Deep Learning-Enabled Real-Time Single-Shot Refocusing of Microwell Array for Digital Melting Curve Analysis
Zhiqi Zhang1,2,3, Jia Yao2,4, Qi Yang2,3
1School of Biomedical Engineering (Suzhou), Division of Life Sciences and Medicine University of Science and Technology of China, Hefei 230026, China.
Analytical Chemistry
|November 4, 2025
Summary
A new deep learning method enhances digital melting curve analysis (dMCA) for multiplexed nucleic acid detection. This advanced platform improves accuracy and resolution, enabling precise gene profiling and advancing precision medicine.
Area of Science:
- Biotechnology
- Molecular Biology
- Computational Biology
Background:
- Digital melting curve analysis (dMCA) is a powerful technique for multiplexed nucleic acid detection.
- Conventional dMCA faces challenges with accuracy across wide temperature ranges and resolution limits due to fluorescence deviations.
Purpose of the Study:
- To develop a novel deep learning-enabled dMCA platform (SAPAR-dMCA) to overcome limitations of conventional dMCA.
- To improve accuracy, resolution, and temperature range adaptability in digital multiplex nucleic acid analysis.
Main Methods:
- Implementation of a single-shot adaptive point-spread function (PSF) attention refocusing model (SAPAR-dMCA).
- Autofocusing of digital PCR microarrays without electromechanical motion using PSF self-calibration and modulation.
- Achieved a ±400 μm depth-of-field, reducing fluorescence intensity deviation by 2.76-fold.
Main Results:
- Significantly enhanced multiplex identification accuracy from 38.0% to 92.3% over a 46.0 °C melting temperature span.
- Reduced coefficient of variation from 3.16% to 0.78%.
- Achieved 0.9 °C resolution for respiratory pathogen detection based on melting temperature differentiation.
Conclusions:
- SAPAR-dMCA offers a precise and robust platform for digital multiplex nucleic acid analysis with high resolution and broad temperature adaptability.
- The developed methodology supports ultramultiplexed gene profiling and advances precision medicine.
