Nanopore basecalling from a perspective of instance segmentation

Abstract

Background: Nanopore sequencing is a rapidly developing third-generation sequencing technology, which can generate long nucleotide reads of molecules within a portable device in real-time. Through detecting the change of ion currency signals during a DNA/RNA fragment's pass through a nanopore, genotypes are determined. Currently, the accuracy of nanopore basecalling has a higher error rate than the basecalling of short-read sequencing. Through utilizing deep neural networks, the-state-of-the art nanopore basecallers achieve basecalling accuracy in a range from 85% to 95%. Result: In this work, we proposed a novel basecalling approach from a perspective of instance segmentation. Different from previous approaches of doing typical sequence labeling, we formulated the basecalling problem as a multi-label segmentation task. Meanwhile, we proposed a refined U-net model which we call UR-net that can model sequential dependencies for a one-dimensional segmentation task. The experiment results show that the proposed basecaller URnano achieves competitive results on the in-species data, compared to the recently proposed CTC-featured basecallers. Conclusion: Our results show that formulating the basecalling problem as a one-dimensional segmentation task is a promising approach, which does basecalling and segmentation jointly.

Genome Sequencing and Analysis of Trichoderma (Hypocreaceae) Isolates Exhibiting Antagonistic Activity against the Papaya Dieback Pathogen, Erwinia mallotivora

“…The large coverage of the long read nanopore sequencing allowed for a robust first draft genome to be created, with few fragments. However, nanopore read base calling remains less accurate than Illumina, around 95% [ 82 ], and while modern base calling software harnessing neural networks are improving this, it is still necessary to polish, with higher accuracy, short read Illumina sequencing data to create a robust genome assembly, in order to remove the indels and miscalled bases present in nanopore reads that otherwise lead to frameshifts and fragmentation of genes.…”

Basecalling Using Joint Raw and Event Nanopore Data Sequence-to-Sequence Processing

“…The Mauler basecaller, proposed by Abbaszadegan [ 13 ], implemented a new approach using deep learning encoder–decoder with the attention model. Even more recent tools are URNano, based on U-net and proposed by Zhang et al [ 14 ], and Causalcall [ 15 ], using TCN, which is a model gaining in popularity. Mentioned tools have increased the accuracy of single strand reads basecalling to over 90–95%.…”

“…As for new conceptions in this field, both Abbaszadegan’s [ 13 ] and Zhang et al’s [ 14 ] works mention that, apart from the direct use of raw data, the processing of its segmentation might be beneficial, providing more information to the model.…”

Applications and potentials of nanopore sequencing in the (epi)genome and (epi)transcriptome era

“…Zhang et al. 40 also proposed a refined U-net model (thus, a UR-net model, an enhanced U-net model for 1D sequence segmentation) called URnano to improve previous end-to-end deep-learning models. Early neural-network base-callers (such as DeepNano 41 and BasecRAWller 42 ) relied on a preprocessing step that segmented the current measurements into discrete events that may artificially introduce errors when segmenting the ion current signals.…”