NanoSim: nanopore sequence read simulator based on statistical characterization

In this work, we model the base-calling errors of ONT reads to inform the simulation of sequences with similar characteristics. We showcase the performance of NanoSim on publicly available datasets generated using the R7 and R7.3 chemistries and different sequencing kits and compare the resulting synthetic reads to those of other long-sequence simulators and experimental ONT reads. We expect NanoSim to have an enabling role in the field and benefit the development of scalable next-generation sequencing technologies for the long nanopore reads, including genome assembly, mutation detection, and even metagenomic analysis software.