Group 8: Challenge: Event detection in nanopore data

Abstract

Translocation of DNA molecules through special membrane nanopores- under the application of electric current- results in change of the nanopore current [1,2]. The produced signals have characteristic patterns that depend on the type of event that takes place, i.e. the configuration of the DNA molecule that passes through the pore. The focus of this challenge is the development of methods for: 1. detecting the events 2. classifying the events based on the signal pattern 3. finding an appropriate measure for quantifying the success of the model The provided data for this challenge consist of nine simulation sets produced by tuning the signal-to-noise ratio (SNR) at 100, 50, 10, 5, 4, 3, 2, 1.5, and 1. Each set (same SNR level) consists of five simulations of electric current time series, resulting in a total of forty-five simulations. An example plot of the data with the maximum SNR of 100 is shown in Figure 1, where the spikes in the data correspond to events. Zooming in on the graph at each of these spikes, it is possible to identify a range of three patterns corresponding to the different classes of event. As the SNR decreases, the true signals become more difficult to distinguish from the natural variation resulting from the noise. Therefore, devising a method to identify events when the SNR is low is a particularly important challenge.

More information

Identifiers

ORCID for Halil Ibrahim Aysel: orcid.org/0000-0002-4981-0827

ORCID for Stephen Gow: orcid.org/0000-0003-0121-1697

ORCID for Jeremy G. Frey: orcid.org/0000-0003-0842-4302

ORCID for Mahesan Niranjan: orcid.org/0000-0001-7021-140X

ORCID for Samantha Kanza: orcid.org/0000-0002-4831-9489

Catalogue record

Export record

Altmetrics