Abdollahzadeh Elnaz, Mortazavi Ali
bioRxiv. 2025 Jun 11:2025.06.04.657941. doi: 10.1101/2025.06.04.657941.
The Oxford Nanopore Technologies (ONT) platform allows for the direct detection of RNA and DNA modifications from unamplified nucleic acids, which is a significant advantage over other platforms. However, the rapid updates to ONT basecalling models and the evolving landscape of computational tools for modification detection bring about challenges for reproducible and standardized analyses. To address these challenges, we developed Dogme, which is a Nextflowbased workflow that automates the processing of ONT data, including basecalling, alignment, modification detection, and transcript quantification. Dogme automates the reprocessing of ONT POD5 files by integrating basecalling using Dorado, read mapping using minimap2 and subsequent analysis steps such as running modkit. The pipeline supports three major types of ONT sequencing data - direct RNA (dRNA), complementary DNA (cDNA), and genomic DNA (gDNA) - enabling comprehensive analyses across different library preparations. Dogme facilitates detection of diverse RNA modifications supported by Dorado such as N6-methyladenosine (m6A), 5-methylcytosine (m5C), inosine, pseudouridine, 2'-Omethylation (Nm) and DNA methylation, while concurrently quantifying full-length transcript isoforms LR-Kallisto for transcript quantification for dRNA and cDNA.
We applied Dogme to three separate mouse C2C12 myoblast replicates using direct RNA sequencing on MinION flow cells. We detected an average of 147,879 m6A, 86,673 m5C, 21,242 inosine, 24,540 pseudouridine, and 83,841 2'- O-methylation sites per replicate with 96,581 m6A, 43,446 m5C, 8,825 inosine, 10,048 pseudouridine, and 30,157 2'-O- methylation sites detected in all three biological replicates. The pipeline produced reproducible modification profiles and transcript expression levels across replicates, demonstrating its utility for integrative long-read transcriptomic and epigenomic analyses.
Dogme is implemented in Nextflow and is freely available under the MIT license at https://github.com/mortazavilab/dogme , with documentation provided for installation and usage.
牛津纳米孔技术（ONT）平台能够直接从未扩增的核酸中检测RNA和DNA修饰，这是相对于其他平台的一个显著优势。然而，ONT碱基识别模型的快速更新以及用于修饰检测的计算工具不断变化的格局给可重复和标准化分析带来了挑战。为应对这些挑战，我们开发了Dogme，这是一个基于Nextflow的工作流程，可自动处理ONT数据，包括碱基识别、比对、修饰检测和转录本定量。Dogme通过整合使用Dorado进行碱基识别、使用minimap2进行读段比对以及后续分析步骤（如运行modkit），实现了ONT POD5文件的重新处理。该流程支持三种主要类型的ONT测序数据——直接RNA（dRNA）、互补DNA（cDNA）和基因组DNA（gDNA）——从而能够对不同文库制备进行全面分析。Dogme有助于检测Dorado支持的多种RNA修饰，如N6-甲基腺苷（m6A）、5-甲基胞嘧啶（m5C）、次黄嘌呤、假尿苷、2'-O-甲基化（Nm）和DNA甲基化，同时使用LR-Kallisto对全长转录本异构体进行定量，用于dRNA和cDNA的转录本定量。
我们在MinION流动槽上使用直接RNA测序将Dogme应用于三个独立的小鼠C2C12成肌细胞重复样本。我们在每个重复样本中平均检测到147,879个m6A、86,673个m5C、21,242个次黄嘌呤、24,540个假尿苷和83,841个2'-O-甲基化位点，在所有三个生物学重复样本中检测到96,581个m6A、43,446个m5C、8,825个次黄嘌呤、10,048个假尿苷和30,157个2'-O-甲基化位点。该流程在重复样本中产生了可重复的修饰图谱和转录本表达水平，证明了其在整合长读长转录组学和表观基因组学分析中的实用性。
Dogme用Nextflow实现，根据MIT许可在https://github.com/mortazavilab/dogme上免费提供，并提供了安装和使用文档。
