Single-molecule sequencing detection of N6-methyladenine in microbial reference materials

Abstract

The DNA base modification N6-methyladenine (m6A) is involved in many pathways related to the survival of bacteria and their interactions with hosts. Nanopore sequencing offers a new, portable method to detect base modifications. Here, we show that a neural network can improve m6A detection at trained sequence contexts compared to previously published methods using deviations between measured and expected current values as each adenine travels through a pore. The model, implemented as the mCaller software package, can be extended to detect known or confirm suspected methyltransferase target motifs based on predictions of methylation at untrained contexts. We use PacBio, Oxford Nanopore, methylated DNA immunoprecipitation sequencing (MeDIP-seq), and whole-genome bisulfite sequencing data to generate and orthogonally validate methylomes for eight microbial reference species. These well-characterized microbial references can serve as controls in the development and evaluation of future methods for the identification of base modifications from single-molecule sequencing data.

Direct RNA nanopore sequencing of full-length coronavirus genomes provides novel insights into structural variants and enables modification analysis

“…In addition, this approach provides insight into the diversity of additional HCoV-229E sg RNAs, probably including DI-RNAs. Further, we aim to assess whether RNA modifications can be called directly from the raw nanopore signal of viral molecules without prior in vitro treatment, as has been shown for DNA (Stoiber et al 2016;McIntyre et al 2019).…”

Recent advances in the detection of base modifications using the Nanopore sequencer

“…They showed that the detection of m5C was easier than that of m6A, contrary to the situation in SMRT sequencing. mCaller can use the four machine learning classifiers (neural network, random forest, logistic regression, and naive Bayes classifiers) to detect m6A on DNA [59]. Mclntyre et al showed that the predictor using the neural network was the most accurate.…”

Methplotlib: analysis of modified nucleotides from nanopore sequencing

“…Over 40 verified types of covalent nucleotide modifications have been described, of which 5-methylcytosine (5mC) and N6-methyladenine (m 6 A) are the most studied (Sood et al , 2019) . The long read sequencing platforms from Oxford Nanopore Technologies (ONT) enable genome-wide direct observation of modified nucleotides by assessing deviating current signals, for which multiple tools have been developed (McIntyre et al , 2019;Liu, Fang, et al , 2019;Liu, Georgieva, et al , 2019;Rand et al , 2017;Stoiber et al , 2016;Simpson et al , 2017) , but a comprehensive evaluation of their performance is lacking. To the best of our knowledge, no flexible genome browser visualization method is available for this type of data.…”