Quantum classifier distinguishes dnmt3a and dnmt3b dna patterns
A 4-Qubit Variational Quantum Classifier for Sequence-Specific Classification of DNMT3A and DNMT3B Flanking Methylation Preferences
Machine Learning
Summary
DNA molecules have chemical markers that help control how genes are turned on or off. Two enzymes, DNMT3A and DNMT3B, place these markers but prefer slightly different DNA patterns, which is hard to tell apart using regular computers. The authors used a small quantum computer simulation to train a classifier that can spot these differences in DNA patterns. Their approach showed promise, suggesting quantum computers might help study complex biological data in the future.
What this means in practice
- •For bioinformatics developers: Create tools for analyzing DNA methylation patterns using quantum machine learning classifiers that differentiate enzyme-specific sequence preferences.
- •For quantum computing engineers: Design and simulate small-scale quantum circuits for proof-of-concept biological data classification tasks, guiding hardware development and software optimization.
Authors
Murat Isik, Yusuf Moinuddin, Humaira Gowher, Sabre Kais
Abstract
DNA methylation is a critical epigenetic modification that regulates gene expression and genome stability, primarily catalyzed by the DNA methyltransferase 3 (DNMT3) family of enzymes. Distinguishing precise sequence-specific flanking preferences between DNMT3A and DNMT3B remains computationally challenging due to the complex and high-dimensional nature of biochemical sequence data. In this study, we propose a novel Quantum Machine Learning (QML) approach to classify DNA methylation preferences, leveraging a Variational Quantum Classifier (VQC). Using experimentally derived deep enzymology data, we constructed a dataset containing 4,096 unique NNNCGNNN motifs. We evaluated a four-qubit VQC using a custom NumPy exact-statevector simulator. The described two-layer circuit contains 24 trainable rotations, eight CNOT gates, and eight input-encoding rotations, totaling 40 gates before measurement. Across five initialization seeds, the revised VQC evaluation achieved a ROC-AUC of 0.661 pm 0.009, where variability denotes sample standard deviation. The held-out test partition is approximately balanced, containing 200 DNMT3A-labeled motifs and 210 DNMT3B-labeled motifs. An always-DNMT3B predictor achieves 51.2% accuracy and a balanced accuracy of 0.500 on this test partition. This proof-of-concept establishes a foundation for utilizing near-term quantum architectures in complex epigenetic classification tasks, paving the way for hybrid quantum-classical pipelines in bioinformatics.