Oxford Nanopore Technology (ONT) Sequencing for the detection of Xanthomonas citri subsp. malvacearum

This analysis workflow was developed to process data generated using an ONT sequencing platform and analyse it for the presence of Xanthomonas citri subsp. malvacearum reads.

Outline

1. Tool requirements and setting up your environment
2. Basecalling
3. Quality assessment
4. Concatenate reads into one file and filter
5. Metagenomic assembly
6. Taxonomic assessment
   • Metamaps
   • Kraken
7. Race differentiation

Unless stated in a code chunk, run code from your chosen working directory. Where the number of threads are specified these can modified as appropriate for your computing envronment.

1. Tool requirements and setting up your environment

This workflow requires a Linux environment and is dependent on the following tools:

• conda
• docker
• Guppy
• Duplex tools
• NanoPlot
• NanoFilt
• metaFlye
• metamaps and/or
• kraken2
• abricate
• krona
• seqkit

Installation

With the exception of guppy and duplex_tools the software dependencies may be managed through conda.

conda create -p Xcm_detection
conda activate Xcm_detection
conda install -c bioconda -c defaults nanoplot
conda install -c bioconda -c defaults nanofilt
conda install -c bioconda -c defaults flye
conda install -c bioconda -c defaults metamaps
conda install -c bioconda -c defaults kraken2
conda install -c bioconda -c defaults krona
conda install -c bioconda -c defaults seqkit
conda install -c bioconda -c defaults abricate

The version of Guppy suitable for your computing environment can be downloaded from the Nanopore Community page. Please follow the relevant installation instructions.

Duplex tools is available from the Nanopore github page. Nanopore recommend running Duplex tools from an isolated virtual environment. Please refer to their installation instructions.

If you have trouble using metamaps with conda, metamaps can be set up on a docker environment:

docker pull nanozoo/metamaps

Navigate to your working directory and create the necessary sub-directories

cd path/directory
mkdir fastq
mkdir fastq_duplex
mkdir nanoplot
mkdir flye
mkdir metamaps
mkdir kraken
mkdir abricate

2. Basecalling

Basecalling can be carried out either using the MinKnow software or with Guppy.

Note here Qscore filtering during basecalling is disabled so quality assessment across the whole dataset can be done. Quality filtering is carried out at a subsequent step. Use the relevant configuration file for your library preparation kit/flow cell combination to run Super Accurate basecalling. The protocol was developed using FLO-MIN106 flow cells/SQK-LSK110 library preparation kit and FLO-MIN114 flow cells/SQK-LSK114 library preparation kit.

FLO-MIN106/SQK-LSK110

guppy_basecaller  --disable_qscore_filtering --input_path directory/reads.fast5 --save_path path/directory -c dna_r9.4.1_450bps_sup -x "cuda:0"

FLO-MIN114/SQK-LSK114

Basecall simplex reads with read splitting enabled.

guppy_basecaller  --disable_qscore_filtering --input_path directory/reads.fast5 --save_path path/directory -c dna_r10.4.1_e8.2_400bps_sup.cfg --do_read_splitting -x "cuda:0"

Identify a list of the template and complement reads using duplex_tools

duplex_tools pairs_from_summary sequencing_summary.txt path/directory
duplex_tools filter_pairs pair_ids.txt path/fastq_directory

Re-basecall using guppy_basecaller_duplex

guppy_basecaller_duplex -i <MinKNOW directory> -r -s duplex_calls -x 'cuda:0' -c dna_r10.4.1_e8.2_400bps_sup.cfg --chunks_per_runner 16 --duplex_pairing_mode from_pair_list --duplex_pairing_file pair_ids_filtered.txt

3. Quality Assessment

Activate the conda environment to proceed with the pipeline.

conda activate Xcm_detection

The quality of the run should be checked using an assessment tool such as Nanoplot (de Coster et al., 2018). Nanoplot is available through the NanoPack package.

The location of the sequencing summary text file varies depending on whether the basecalling was post sequencing or during. You MUST modify this code chunk directing it to the sequencing summary file relevant to your dataset.

NanoPlot -t 10 --summary directory/fastq/sequencing_summary.txt -p SampleName -o nanoplot

If the run quality is low, preparing and sequencing another library is recommended.

4. Concatenate all reads into one file and filter with NanoFilt

Concatenate read files into a single fastq file and filter using a tool such as Nanofilt (de Coster et al., 2018).

cat directory/*.fastq > directory/all_SampleName_raw.fastq

We recommend using different filtering criteria depending on the flowcells used.

FLO-MIN106/SQK-LSK110

Remove read less than 1000bp or with a phred quality < 7 (default)

cat directory/all_SampleName_raw.fastq | NanoFilt -l 1000 -s fastq/sequencing_summary.txt > directory/SampleName_filtered_l1000.fastq

FLO-MIN114/SQK-LSK114

Remove read less than 1000bp or with a phred quality < 8 (default)

cat directory/all_SampleName_raw.fastq | NanoFilt -l 1000 -q 8 -s fastq/sequencing_summary.txt > directory/SampleName_filtered_l1000.fastq

5. Metagenomic Assembly

Assemble the metagenome using a long-read metagenome assembler tool such as metaFlye (Kolmogorov et al., 2020).

flye --nano-raw directory/SampleName_filtered_l1000.fastq --genome-size 6m --out-dir directory/flye --threads 20 --meta

6. Taxonomic assessment

MetaMap Analysis

The taxonomic profile of reads can be assigned using a tool such as metamaps (Dilthey et al., 2019)

Here the miniSeq+H database (downloaded 16/10/2020) was used.

metamaps  mapDirectly -t 40 --all -r directory/metamaps/databases/databases/miniSeq+H/DB.fa -q directory/SampleName_filtered_l1000.fastq -o directory/metamaps/classification_results

metamaps classify --mappings directory/metamaps/classification_results --DB directory/metamaps/databases/databases/miniSeq+H -t 40

Filter taxonomic assignments for a minion of 80% percent identity

perl path/MetaMaps-master/util/filterLowIdentityEntities.pl --DB miniSeq+H --mappings directory/metamaps/classification_results --identityThreshold 0.8
Rscript plotMappingSummary.R directory/metamaps/classification_results_filt80

Visualise using krona (Ondov et al., 2011).

ktImportTaxonomy -i -m 2 -o directory/metamaps/SampleName.html directory/metamaps/classification_results_filt80.EM.reads2Taxon.krona

Extract reads classified as Xcm for further analysis using seqkit (Shen et al., 2016).

awk '{if ($2 == "86040") print $0;}' metamaps/classification_results.EM.reads2Taxon | cut -f 1 > Xcm1.ids
wc -l Xcm1.ids

awk '{if ($2 == "1118965") print $0;}' metamaps/classification_results.EM.reads2Taxon | cut -f 1 > Xcm2.ids
wc -l Xcm2.ids

awk '{if ($2 == "1127439") print $0;}' metamaps/classification_results.EM.reads2Taxon | cut -f 1 > Xcm3.ids
wc -l Xcm3.ids

awk '{if ($2 == "1220027") print $0;}' metamaps/classification_results.EM.reads2Taxon | cut -f 1 > Xcm4.ids
wc -l Xcm4.ids

awk '{if ($2 == "1220028") print $0;}' metamaps/classification_results.EM.reads2Taxon | cut -f 1 > Xcm5.ids
wc -l Xcm5.ids

Combine the individual files into one file containing the read names for Xcm reads.

cat Xcm1.ids Xcm2.ids Xcm3.ids Xcm4.ids Xcm5.ids > Xcm_all.ids

Double check that the total number of reads in Xcm_all.ids matches the sum of the individual files from above:

wc -l Xcm_all.ids

Extract the Xcm reads from filtered data using seqkit

seqkit grep -f Xcm_all.ids directory/SampleName_filtered_l1000.fastq -o SampleName_Xcm_reads_metamaps.fastq

Kraken

Alternatively Kraken2 (Wood et al., 2019) can be used to analyse the raw reads or the metaFlye assembly. Details on building a kraken2 database from genomes in RefSeq can be found at: https://github.com/DerrickWood/kraken2/blob/master/docs/MANUAL.markdown

Kraken2 can be run on either the metaflye assembly or the filtered reads.

Kraken2 on the metaflye assembly:

kraken2 --db directory/kraken2db/Refseq91/ --threads 20 --unclassified-out directory/kraken/SampleName_kraken2fl.unclassified.fasta --report directory/kraken/SampleName_kraken2fl.report --output directory/kraken/SampleName_kraken2fl.txt directory/flye/assembly.fasta

Kraken2 on the filtered reads:

kraken2 --db directory/kraken2db/Refseq91/ --threads 20 --unclassified-out directory/kraken/SampleName_kraken2fl.unclassified.fasta --report directory/kraken/SampleName_kraken2fl.report --output directory/kraken/SampleName_kraken2fl.txt directory/SampleName_filtered_l1000.fastq

Visualise using krona

ktImportTaxonomy -m 3 -t 5 -o directory/kraken/SampleName_kraken2f1.html directory/kraken/SampleName_kraken2fl.report

7. Race differentiation

Search for race specific genes using Abricate (Seemann). A database has to be set up for this to work.

Make a database using the customized fasta file Race18_abricate.fasta provided here.

cd directory/abricate/db
cp directory/Race18_abricate.fasta sequences
makeblastdb -in sequences -title race_specific -dbtype nucl -hash_index

Use abricate to search for the sequences in Race18_abricate.fasta

abricate --db race_specific flye/assembly.fasta

Acknowledgements

This project is supported by the Grains Research and Development Corporation, through funding from the Australian Government Department of Agriculture, Fisheries & Forestry, as part of its Rural R&D for Profit program and along with Cotton Research and Development Corporation, Hort Innovation Australia, Wine Australia, Sugar Research Australia and Forest and Wood Products Australia.

References

1. De Coster W, D’Hert S, Schultz DT, Cruts M, Van Broeckhoven C. NanoPack: visualizing and processing long-read sequencing data. Bioinformatics. 2018;34(15):2666-9.
2. Dilthey AT, Jain C, Koren S, Phillippy AM. Strain-level metagenomic assignment and compositional estimation for long reads with MetaMaps. Nature Communications. 2019;10(1):3066.
3. Kolmogorov M, Bickhart DM, Behsaz B, Gurevich A, Rayko M, Shin SB, et al. metaFlye: scalable long-read metagenome assembly using repeat graphs. Nature Methods. 2020;17(11):1103-10.
4. Ondov BD, Bergman NH, Phillippy AM. Interactive metagenomic visualization in a Web browser. BMC Bioinformatics. 2011;12(1):385.
5. Seemann T, Abricate, Github https://github.com/tseemann/abricate
6. Shen W, Le S, Li Y, Hu F. SeqKit: A Cross-Platform and Ultrafast Toolkit for FASTA/Q File Manipulation. PloS one. 2016;11(10):e0163962.
7. Wood DE, Lu J, Langmead B. Improved metagenomic analysis with Kraken 2. Genome Biology. 2019;20(1):257.