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

Abstract

Third-generation DNA sequencers provided by Oxford Nanopore Technologies (ONT) produce a series of samples of an electrical current in the nanopore. Such a time series is used to detect the sequence of nucleotides. The task of translation of current values into nucleotide symbols is called basecalling. Various solutions for basecalling have already been proposed. The earlier ones were based on Hidden Markov Models, but the best ones use neural networks or other machine learning models. Unfortunately, achieved accuracy scores are still lower than competitive sequencing techniques, like Illumina’s. Basecallers differ in the input data type—currently, most of them work on a raw data straight from the sequencer (time series of current). Still, the approach of using event data is also explored. Event data is obtained by preprocessing of raw data and dividing it into segments described by several features computed from raw data values within each segment. We propose a novel basecaller that uses joint processing of raw and event data. We define basecalling as a sequence-to-sequence translation, and we use a machine learning model based on an encoder–decoder architecture of recurrent neural networks. Our model incorporates twin encoders and an attention mechanism. We tested our solution on simulated and real datasets. We compare the full model accuracy results with its components: processing only raw or event data. We compare our solution with the existing ONT basecaller—Guppy. Results of numerical experiments show that joint raw and event data processing provides better basecalling accuracy than processing each data type separately. We implement an application called Ravvent, freely available under MIT licence.

Portable nanopore-sequencing technology: Trends in development and applications

“…EpiNano, an algorithm for predicting m6A RNA modification from dRNA sequence data sets, can now train models with features extracted from both basecalled dRNA seq FASTQ data and raw FAST5 nanopore outputs ( Liu H. et al, 2021 ). Ravvent is a new basecaller that uses joint processing of raw and event data and is based on an encoder-decoder architecture of recurrent neural networks ( Napieralski and Nowak, 2022 ). NanoReviser, an open-source DNA basecalling reviser, is based on a deep learning algorithm to correct the basecalling errors introduced by current basecallers provided by default ( Wang et al, 2020 ).…”

A full‐length 18S ribosomal DNA metabarcoding approach for determining protist community diversity using Nanopore sequencing

“…At present, ONT has high accuracy basecalling tools that has significantly reduced the error rate, for example, guppy, buttery‐eel, and dorado (Gamaarachchi et al., 2022 ). Furthermore, Nanopore basecalling has been in continuous development in recent years to improve sequence accuracy (Napieralski & Nowak, 2022 ; Pagès‐Gallego & de Ridder, 2023 ). Additional advantages of the ONT include the following: it is cost effective, its small size makes it easily transportable, and the library preparation for sequencing is relatively simple (van der Reis et al., 2023 ; Wang et al., 2021 ).…”