MrHAMER2 is a pipeline for generating of high accuracy single molecule Nanopore reads. The pipeline accepts FASTQ-format sequence files as input and outputs a multi-FASTQ of corrected non-chimeric reads, aligned reads and QC stats.
This is a heavily customized and enhanced version of https://github.com/nanoporetech/pipeline-umi-amplicon from Oxford Nanopore Technologies.
The pre-print is available at https://www.biorxiv.org/content/10.1101/2024.12.19.629526
The pipeline performs the following steps:
- Reads are mapped to reference genome using minimap2
- Separate into amplicons
- Extract UMI sequences for all reads
- Cluster UMI sequences per amplicon using vsearch and compute high accuracy consensus reads
- Remove PCR-derived chimeras using Chimera_Buster (optional)
This pipeline has only been tested on Ubuntu 20.04.5 LTS.
The following software packages must be installed prior to running:
- miniconda3 - please refer to installation instructions.
After installing miniconda3, install the pipeline as follows:
git clone https://github.com/gallardo-seq/MrHAMER2Change to directory:
cd MrHAMER2Create conda environment with all dependencies:
conda env create -f environment.ymlActivate environment:
conda activate MrHAMER2Install Auxiliary Python Scripts
cd lib && pip install . && cd ..Copy modified Medaka files provided by MrHAMER2 to conda environment:
cp medaka_mod/* $CONDA_PREFIX/lib/python3.8/site-packages/medakaDeactivate environment:
conda deactivateTo test if the installation was successful run:
conda activate MrHAMER2
snakemake -j 4 --configfile config.ymlIn order for the pipeline to run, the following files/folders must be in the working directory:
| File | Location in MrHAMER2 directory |
|---|---|
| Snakefile | ./Snakefile |
| config.yml | ./config.yml |
To run the pipeline the following input files are required:
| Input | Description |
|---|---|
| Long-read sequences | Folder containing FASTQ files or a single concatenated FASTQ file. |
| Reference genome | FASTA file containing the reference genome. |
| Target BED file | A BED file containing the chromosome, start and end coordinate and the name of all amplicons. |
| Medaka Model file | A tar.gz file that provides the model for Medaka processing. NOTE: The medaka model file for R10.4 chemistry with e8.2 pores at 400 bps speed is provided. For more information on Medaka models, see medaka model info. If a different model is needed, please download from medaka model download. |
Before starting the pipeline, confirm the following varibles are correct in the config.yml file:
| Input | Description |
|---|---|
| sample_name | The name that the output folders/files will use. |
| input_fastq | The location of the long-read sequences fastq file. |
| reference_fasta | The location of the reference genome fasta file. |
| targets_bed | The location of the target BED file. |
| You can also adjust the optional parameters listed below: | |
| Parameter | Description |
| ------- | ------------- |
| chimera_buster_on | If set to true, the Chimera_Buster module will be run at the end of the pipeline and will output a fastq file of nonchimeric corrected reads. |
| chimera_buster_MT | Designates the maximum number of mismatched/indel bases allowed when comparing UMIs. Changing this value will change the specificity and throughput of the Chimera_Buster module. Default is 1. |
| chimera_buster_HS | When set to true, chimeric reads are rechecked to account for any clustering issues earlier in the pipeline. WARNING: this part of the code is slow and is only recommended for low input samples. Default is False. |
| minimap2_param | This sets the parameters that Minimap2 will use throughout the pipeline. Please refer to the manual for Minimap2 linked above for more details on the availble parameters. |
| umi_errors | Designates the maximum differences between UMI in read and UMI pattern. |
| min_reads_per_cluster | Minimum number of reads required for a consensus read. |
| max_reads_per_cluster | Maximum number of 1D reads used for a consensus read. |
| min_overlap | Minimum overlap with target region. |
| balance_strands | If set to True, the pipeline inforces balancing of forward and reverse 1D reads in clusters. |
| medaka_model | Location of Medaka model used to compute consensus reads. |
| fwd_context | Sequence of tail on 5' end of the forward UMI primer. |
| rev_context | Sequence of tail on 5' end of the reverse UMI primer. |
| fwd_umi | Pattern of the UMI in the forward UMI primer. |
| rev_umi | Pattern of the UMI in the reverse UMI primer. |
| min_length | Minimum combined UMI length. |
| max_length | Maximum combined UMI length. |
Once the config.yml is updated with the correct information and the MrHAMER2 environment is activated, simply run the following command:
snakemake -j 32 --configfile config.yml-j specifies how many threads will be used by the pipeline.
The main output file created by the pipeline is:
| Output | Description |
|---|---|
| {sample_name}_{target}_final.fastq | A multi-fastq of all corrected non-chimeric reads. |
After the a pipeline analysis has completed, these files can be found at {sample_name}/{sample_name}_{target}_final.fastq e.g. MrHAMER2_test/MrHAMER2_test_Env_final.fastq.
For issues or bugs, please report them on the issues page.
MIT - Copyright (c) 2024 Christian M Gallardo and Jessica L Albert