Related Experiment Video
Updated: Jun 10, 2025
07:16
Author Spotlight: Advancements in DNA Nanosensors – Addressing Sensitivity and Selectivity Challenges in Molecular Detection
Published on: February 9, 2024
899
Applying Machine Learning and SERS for Precise Typing of DNA Secondary Structures.
Guantong Xu1, Ying Bao2, Yujing Zhang1
1State Key Laboratory of Supramolecular Structure and Materials, College of Chemistry, Jilin University, 2699 Qianjin Street, Changchun 130012, P. R. China.
Analytical Chemistry
|October 16, 2024
Summary
This study combines machine learning with surface-enhanced Raman spectroscopy (SERS) to accurately identify DNA secondary structures like G-quadruplex (G4) and i-motif (iM). The new method enables rapid screening and prediction of various DNA conformations.
Area of Science:
- Biophysical Chemistry
- Molecular Biology
- Computational Biology
Background:
- Surface-enhanced Raman spectroscopy (SERS) is effective for analyzing DNA secondary structures.
- Challenges exist in identifying complex DNA conformations due to spectral overlap and limited sample data.
Purpose of the Study:
- To develop a synergistic machine learning and SERS approach for accurate DNA secondary structure identification.
- To analyze SERS spectra of diverse DNA conformations including G-quadruplex (G4), i-motif (iM), double-strand (DS), and single-strand (SS).
Main Methods:
- Utilized Principal Component Analysis (PCA) for initial conformational grouping.
- Applied Linear Discriminant Analysis (LDA), K-Nearest Neighbor (KNN), and Support Vector Machine (SVM) for classification.
- Analyzed SERS spectra from 54 well-defined oligonucleotides.
Main Results:
- PCA successfully segregated oligonucleotides into G4, iM, and other groups.
- Machine learning models achieved high accuracy in classifying trained sequences.
- Correctly identified structures of five untrained sequences and predominant conformations (G4, iM, DS) under varying pH conditions.
Conclusions:
- The integrated SERS and machine learning methodology shows significant potential for rapid DNA secondary structure screening.
- This approach can accurately predict and identify complex DNA conformations, aiding in molecular diagnostics and research.
