Synthetic and experimental nanopore signals for sublevel event detection and denoising using decision trees.

This dataset contains experimental and synthetic nanopore signal data to support signal processing research, including event classification and sub-level fitting. The experimental dataset consists of real nanopore recordings obtained using a Chimera VC100 instrument at a sampling rate of 4.167 MHz. This instrument captured the translocation of DNA nanostructures with distinguishable levels of blockage and encoded structural variations in the molecules. The data were recorded in 3.6 M LiCl at 150 mV bias using a 14.2 nm solid state nanopore in a 10 nm thick SiN membrane and low pass filtered at 250 kHz using an 8 pole low pass Bessel filter. The synthetic data set was generated using Python's NumPy and SciPy libraries. We generated a piecewise constant signal and distorted it using a response function for nanopore measurements. We then added zero-mean white noise and sampled the data at 5Mhz and low-pass filtered to 1Mhz using a 4-pole zero-phase low-pass Bessel filter. This research was funded by the Natural Sciences and Engineering Research Council of Canada (NSERC), the Ontario Centre of Innovation (OCI) and Oxford Nanopore Technologies.