DNA extraction and library preparation for rapid genus- and species-level identification, with or without PCRCombining rapid DNA extraction and library-preparation methods with real-time analysis workflows enables unbiased identification of species with unparalleled speed

Fig. 1 Identification of bacterial, archaeal, fungal and viral species using rapid gDNA library preparation and the real-time WIMP analysis workflow

As a strand of DNA from an organism passes through a nanopore, the electrical current flowing through the pore is measured, and these current levels are converted into basecalls in real time. The basecalled read can then be compared to a database of reference genomes using the What’s in My Pot? (WIMP) workflow, also in real time, and the species identified. Real-time data analysis gives us the potential to perform very rapid species identification. A major rate-limiting step in this process can be the extraction of sufficiently pure, high-molecular weight DNA from the sample, and so to address this limitation we evaluated several different methods of cell lysis, including both enzymatic and physical approaches. We found that ClaremontBio’s OmniLyse Lysis kit gave a good combination of rapid lysis while being able to maintain reasonably high fragment length. This lysis method employs bead-beating to disrupt cells from a wide range of organisms and can be performed in 1–2 minutes. DNA extracted in this way is contaminated with cell debris, and needs to be purified in order to perform successful library preparation. DNA of sufficient purity can be quickly obtained by passing the cell lysis extract through a spin column. Although columns of this type typically remove DNA fragments greater than around 10 kb, the vast majority of strands produced by bead beating with the OmniLyse fall below this size limit and are retained in the eluate. Alternatively, a SPRI bead clean-up can be performed. Following elution, we prepared a sequencing library in 5 minutes using our rapid 1D gDNA kit, and we loaded the library onto a flowcell. The first reads were analysed by the WIMP workflow shortly after the run was started and the species identified, meaning that we can go from sample to identification of the species in under 20 minutes (Fig. 1).

Extraction, library prep and species ID in under 20 minutes by whole-genome sequencing

Fig. 2 Identification of species using 16S PCR with the rapid amplicon kit, and the rapid whole-genome amplification (WGA) protocol

In situations where the amount of available sample DNA is limited, or where there is a low level of pathogen DNA mixed with a high level of host DNA, and we wish to identify the pathogen, it can be helpful to amplify the target organism by PCR. For bacterial species identification, the 16S ribosomal RNA gene can be amplified from all bacteria non-specifically, without amplifying eukaryotic host DNA, or viruses. As with the whole-genome species ID approach shown in Fig. 1, we have found bead-beating to lyse cells rapidly, yielding DNA with a sufficiently high fragment length for amplification of the 1.5 kb 16S gene. There is no need to purify the extracted DNA before PCR. Fast polymerases are available which can process 1 kb of template in around 30 seconds, meaning that 16S PCR can be performed on a standard thermocycler in 25 minutes. If PCR is performed using Oxford Nanopore’s modified primers, sequencing adapters can be attached rapidly following amplification, by chemical ligation. This reaction tends to be more robust than enzymatic ligation, so again, there is no need to perform a clean-up step before adapter attachment. Finally, the sequencing library is loaded onto a MinIONTM flowcell and the run started. Using the 16S analysis workflow, species identities are returned in almost real time, making it possible to go from cells to identification in under 40 minutes (Fig. 2). Alternatively, if all the genomes present in a small amount of sample are to be sequenced and analysed, it is possible to amplify the sample by WGA. This can generate sufficient DNA for sequencing in 90 minutes, starting with picogram quantities of input. Following this reaction, adapters are added using the rapid genomic DNA library kit, and the resulting library is loaded and sequenced. The WIMP workflow can then be used to identify species from the resulting whole genome data, with the first results being generated in under 100 minutes from the start of the workflow.

© 2017 Oxford Nanopore Technologies. All rights reserved.P17007 - Version 4.0

PCR-based species ID in under 40 minutes and WGA-based species ID in under 100 minutes

Attachment ofsequencingadapters

PCR total <40 minutes

Adapter attachment: 5 minutes Analysis: real time

Staphylococcus

Bacillus

Listeria

Enterococcus

Lactobacillus

Salmonella

Escherichia

Shigella

Klebsiella

Enterobacter

Alignment countover 80% accuracy

0 10,000 20,000 30,000 40,000

Species identification – key figures

Selection summary

of 16S gene – uses most accurate classification of each read

Top classifications

139329Reads analysed

139237Classification

1320Unique taxa

Staphylococcus

Distribution of alignment accuracies

80 90 1000

1,000

2,000

3,000

Alignment accuracy

Lineage

NCBI Taxonomy ID:

Rank:

Average alignment accuracy:

Alignments at this node:

Alignments (including child nodes):

1279NCBI organism overviewNCBI taxonomy overviewgenus

88.5 %

0

40477

superkingdom:phylum:

class:order:family:

BacteriaFirmicutesBacilliBacillalesStaphylococcaceae

genus: Staphylococcus

16S BLASTN report

Extraction: 2 minutes

Sample

Resuspension in buffer

Cell lysis bybead-beating(e.g. OmniLyse)

PCRamplification

Bead-washing and elution

FR

16S PCR: 30 minutes

WGA:~90 minutes

WGA

Transposomecomplex

WGA total <100 minutes

Purified high-molecular weight gDNA

Clean-up: 5 minutes

High-molecularweight gDNA

Transposome complex (transposase + adapters)

Cleavage and addition oftransposase adapters

Attachment of 1Dsequencing adapters Total

<20 minutes

Column or AMPure/SPRI clean-up

Library prep: 5 minutes Analysis: real time

NCBI taxonomy

Taxonomic lineage (NCBI)Filter by read count0 10011

Aspergillusfumigatus Af293

Tax ID:

Rank:

Score:

Read Count:

330879

strain

0.0870

1457* Arc angle is proportional to its read count

Sample composition Selection

Highest Score0.0913

Lowest Score0.0234

Taxonomic lineage (NCBI)

- Eukaryota-- Fungi--- Dikarya---- Ascomycota----- Pezizomycotina------ Eurotiomycetes------- Eurotiomycetidae

--------- Aspergillus

superkingdom

phylumclasssubclass

kingdomsubkingdom

orderfamilygenusspecies

Taxonomic lineage (NCBI)

---------- Aspergillus fumigatus

-------- Eurotiales

Taxonomic lineage (NCBI)Threshold

Score0.040.020.00 0.06 0.08 0.10

Filter by classification score

28318reads

Neisseria gonorrhoeae

Bordetella pertussis

Peptoclostridium difficile

Human parainfluenza virus 1

Mumps virus

Measles virus

Human coronavirus 229E

Campylobacter jejuni

Helicobacter pylori

Salmonella enterica subsp. enterica

Treponema denticola ATCC 35405

Respiratory syncytial virus

Akkermansia muciniphila ATCC BAA-835

Malassezia globosa CBS 7965

Cryptococcus neoformans JEC21

Candida albicans SC5314

Aspergillus fumigatus Af293

Penicillium marneffei

Extraction: 2 minutes

Sample

Resuspension in buffer

Cell lysis bybead-beating(e.g. OmniLyse)

Contact: publications@nanoporetech.com More information at: www.nanoporetech.com and publications.nanoporetech.com