Dilrini De Silva, a Technical Applications Scientist in Bioinformatics at Oxford Nanopore Technologies, spoke about Methylation Detection on MinKNOW as part of the Human genome sequencing and analysis Nanopore Learning course. They discussed approaches for detecting methylation states with Nanopore sequencing and the MinKNOW software. MinKNOW includes models for methylation detection and mapping of reads. MinKNOW uses Guppy with Remora and Minimap2 and produces a Bam output file. With Nanopore technology, detection of methylation is more reproducible than other methods, De Silva noted. MinKNOW can perform methylation detection while basecalling. Canonical base calling generates fastq files and methylation detection runs in parallel. If you provide a reference genome, alignment is performed. Downstream applications like the human variation workflow, can accept aligned Bam format. You can also perform post-run methylation detection using Guppy. For this, you need to supply input and output files and a configuration file. These files are typically found in a subdirectory. The process is computationally intensive. Methylation detection can be performed with the human variation workflow in EPI2ME Labs that takes Pod5 or Fast5 files as input. An HTML report is created. I am interested in learning more about methylation detection for some of the genomes we are sequencing.
