About
Dorado is a high-performance, open-source analysis engine developed by Oxford Nanopore Technologies (ONT) for processing raw nanopore sequencing reads into usable sequence data. As the next-generation replacement for ONT's legacy Guppy basecaller, Dorado is designed for researchers and bioinformaticians who need fast, accurate, and flexible basecalling capabilities. The tool runs as a single executable with sensible defaults and automatic hardware detection, making it straightforward to deploy in diverse computing environments. It supports Apple Silicon (M-series) and NVIDIA GPUs with linear multi-GPU scaling for high-throughput workflows. For CPU-only environments, Dorado still operates but with reduced throughput. Dorado supports simplex and duplex basecalling modes. Duplex mode leverages both strands of a DNA molecule for ultra-high accuracy reads. The engine also natively performs modified basecalling, enabling direct detection of DNA base modifications such as 5-methylcytosine (5mC) and 6-methyladenine (6mA) without separate post-processing steps. Simplex barcode classification supports demultiplexing of pooled sequencing libraries. Output is provided in industry-standard SAM/BAM alignment formats, integrating seamlessly into downstream bioinformatics pipelines. Dorado is ideal for whole-genome sequencing, epigenomics, transcriptomics, and metagenomics research on ONT platforms including MinION, PromethION, and GridION.
Key Features
- GPU-Accelerated Basecalling: Runs on Apple Silicon (M-series) and NVIDIA GPUs with linear scaling across multiple GPUs for maximum throughput.
- Modified Basecalling: Directly detects DNA base modifications such as methylation (5mC, 6mA) during the basecalling step, eliminating separate post-processing.
- Duplex Basecalling: Leverages both strands of a double-stranded DNA molecule to produce ultra-high accuracy consensus reads.
- Simplex Barcode Classification: Built-in demultiplexing of barcoded, multiplexed sequencing libraries without requiring additional tools.
- SAM/BAM Aligned Output: Natively outputs aligned reads in industry-standard SAM/BAM format for seamless integration with downstream bioinformatics pipelines.
Use Cases
- Whole-genome sequencing analysis of organisms using Oxford Nanopore sequencing platforms such as MinION or PromethION.
- Epigenomics research requiring direct detection of DNA base modifications like 5mC methylation without additional post-processing tools.
- High-throughput multiplexed sequencing experiments requiring barcode demultiplexing of pooled libraries.
- Ultra-high accuracy sequencing projects using duplex basecalling to generate consensus reads from both DNA strands.
- Metagenomics and transcriptomics workflows needing fast, GPU-accelerated basecalling integrated into bioinformatics pipelines.
Pros
- Fully Open Source: Freely available and actively maintained by Oxford Nanopore Technologies on GitHub, with transparent development and community contributions.
- All-in-One Analysis Engine: Handles basecalling, base modification detection, duplex processing, and barcode classification in a single tool with sensible defaults.
- Multi-Platform GPU Support: Supports both Apple Silicon and NVIDIA GPUs with multi-GPU linear scaling, making it suitable for laptops through HPC clusters.
- Automatic Hardware Detection: Automatically detects and configures available hardware, lowering the barrier to entry for new users.
Cons
- GPU Required for Practical Performance: While CPU execution is possible, real-time or high-throughput basecalling requires a compatible GPU, which may limit accessibility.
- ONT-Specific Tool: Designed exclusively for Oxford Nanopore sequencing data and is not compatible with other sequencing platforms such as Illumina or PacBio.
- Command-Line Only: Dorado has no graphical user interface, requiring familiarity with the command line and bioinformatics workflows.
Frequently Asked Questions
What is basecalling in nanopore sequencing?
Basecalling is the process of converting raw electrical signals produced by a nanopore sequencer into DNA or RNA sequence strings (A, T, C, G bases). Dorado uses deep learning models to perform this conversion accurately.
What hardware does Dorado support?
Dorado supports Apple Silicon (M-series chips) and NVIDIA GPUs, including multi-GPU setups with linear performance scaling. CPU-only execution is supported but is significantly slower.
Is Dorado free to use?
Yes. Dorado is fully open source and free to use. It is hosted on GitHub under an open-source license by Oxford Nanopore Technologies.
What is the difference between simplex and duplex basecalling?
Simplex basecalling processes a single strand of DNA, while duplex basecalling uses both the template and complement strands of a double-stranded DNA molecule to produce a higher-accuracy consensus read.
How does Dorado differ from Guppy?
Dorado is ONT's next-generation basecaller, designed to replace the legacy Guppy tool. It offers improved performance, broader GPU support, native base modification detection, and a more unified feature set in a single executable.
