Answers are grounded in this paper and cite their sources.
Nanopores enable single-molecule analysis by measuring current signals through nanoscale pores in either biological or solid-state membranes. Accurate detection of analyte fingerprints within the pore environment is essential for reading-out the analyte type and sequence. Towards these aims, we develop a framework for robust, semi-supervised detection of molecular nanopore events on the basis of a graph representation of the measured signals. To this end, using experimental signals from DNA translocation through an aerolysin pore, we build a graph-based two-stage workflow employing convolutional and graph neural networks that first perform fast screening followed by deep validation of these nanopore events. The learned model is able to select possible molecular signatures (the current blockades) in the full signal, denoise, validate, and reconstruct them, and predict the morphology of unseen molecular events within the range of the available data. The trained model can efficiently predict the correct event morphology for the same analyte within a 2.4-fold range of transmembrane voltage values not included in the training. The developed partially supervised workflow, linking unsupervised graph representation-learning to raw-signal inference is - at its current state - a computational proof of concept that proposes a direction for the development of modular and generalizable event detection workflows, provided that they are trained on a huge amount of different nanopore experiments.
No takes yet. Share an insight, caveat, or question.
Kansari et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: