artic-doc

Measles resources

Thu, 12 Feb 2026 00:00:00 +0000

As part of the Wellcome Trust funded ARTIC 2 project we have developed a collection of measles virus resources that are now openly available at artic.network/viruses/mev. These resources provide guidance for public health and research labs who are planning to perform genome sequencing and analysis of measles virus. We have assembled an end‑to‑end package including primers and sequencing resources, best practice pipelines for sequence data processing, and guidance for analysis and interpretation.

Optimisation of SMART-9N for viral metagenomics

Mon, 08 Sep 2025 00:00:00 +0000

Metagenomics enables the analysis of the entire nucleic acid contents of a sample without target-specific amplification. Therefore, it presents itself as an untargeted strategy for viral surveillance, allowing the detection of both known and emerging species and variants. However, sensitivity proves challenging due to the complex microbiomes of samples and high host nucleic acid backgrounds. To overcome these setbacks, host nucleic acid depletion can be deployed, and viral RNA and DNA can be amplified.

For RNA virus detection and sequencing, SMART-9N has been utilised successfully, recovering 10 kb of the Zika virus genome in a single sequence read. SMART-9N builds on the SMART (Switching Mechanism at the 5′ end of RNA Template) concept, utilising 9N primers to randomly prime RNA for cDNA synthesis. Traditionally, SMART targets polyadenylated RNA, however, the 9N primer utilised in SMART-9N binds independently of the poly(A) tail. During reverse transcription, non-templated cytosines are added at the 3’ end of the cDNA strand. SMART-9N utilises a template-switching primer containing riboguanosines to bind to the cytosines allowing the reverse transcriptase enzyme to add the primer sequence to the 5’ end. Therefore, both ends of the cDNA contain the specific primer sequence that can be used to amplify the cDNA and enrich for RNA viruses without virus-specific primers. Furthermore, as the primers utilised are independent of the poly(A) tail of RNA, they can also anneal to and amplify DNA in a similar mechanism as described in SISPA-Seq. Hence, SMART-9N facilitates unbiased, high-resolution metagenomic sequencing for both DNA and RNA viruses.

Recently, the SMART-9N protocol has been optimised and updated to ensure that the methodology utilised produces the best results for viral metagenomics. The following optimisations have been implemented:

Host Depletion and Extraction:

The original protocol involves a spin-column extraction followed by DNase treatment and a clean-up column.
The optimised protocol performs DNase treatment first, followed by a magnetic bead extraction.
This facilitates the removal of extracellular DNA present in the sample and allowing DNA viruses to be processed through the extraction and hence, amplified in the SMART-9N reaction.

Primer Concentration:

Previously, the original protocol utilised a primer concentration of 2 µM for annealing and cDNA synthesis, followed by 20 µM of PCR primer.
The optimised protocol utilises a primer concentration of 12 µM for annealing and cDNA synthesis, followed by 10 µM of PCR primer.
The amended concentrations produced a greater yield and genome coverage in comparison to the original concentrations.

Rapid SMART-9N:

The original protocol utilised the ONT RLB barcoding primers (SQK-RPB114.24) to perform amplification and barcoding in a single reaction.
However, the ONT RLB barcoding primer fails to sufficiently amplify the cDNA, in comparison to an unmodified PCR primer that produces a ten-fold greater yield.
Therefore, the optimised protocol utilises the unmodified PCR primer and requires separate library preparation to barcode and adapt the samples.

The updated protocol is published on protocols.io: doi.org/10.17504/protocols.io.rm7vz9xn5gx1/v1

The ZeptoMetrix® NATtrol™ Respiratory Panel 2.1 (RP2.1), containing 18 viral and 4 bacterial taxa, was utilised to compare the original SMART-9N protocol to the optimised version, subsampled to 10 gbp. K-mer containment, a useful metric to assess the completeness and accuracy of assembled genomes, was used to compare the methods directly. Overall, the optimised protocol proves to recover a greater containment of the respiratory genomes present in the control.

Are sewers a gold mine of epidemiological data? Applying the ARTIC approach to wastewater

Thu, 14 Aug 2025 00:00:00 +0000

What is wastewater-based epidemiology?

Interest in wastewater-based epidemiology (WBE) peaked during the SARS-CoV-2 pandemic; a number of governments turned to it in the hope of informing their strategy to manage the virus at a time when balancing economic needs with protecting public health was a top priority. But what is WBE and what does it have to do with the ARTIC project?

WBE, as its name suggests, consists in the use of wastewater to track the health of populations, including the prevalence of viral diseases. For this it relies on the fact that many viruses, even respiratory ones, are often shed in the faeces of infected people. Because sewers collect faeces from entire populations, the amount of a virus that is detected in a sewer can serve as a measure of the prevalence of that virus in the population it serves.

WBE has several advantages over clinical testing, and perhaps the most notable is how efficient it is. To provide a representative picture of infection burden of a population through clinical testing, a representative number of samples must be taken and processed, involving substantial efforts and cost; on the other hand, a single sample of wastewater can provide an estimate for the entire population within a sewershed (the area drained by one discrete sewer system). Because everyone contributes to the production of wastewater, this approach does not miss asymptomatic cases. Furthermore, in the case of SARS-CoV-2 it has been found on several occasions (Karthikeyan et al. 2022; Nemudryi et al., 2020) that wastewater monitoring can identify epidemiological trends earlier than clinical testing. These advantages were what caused interest in WBE to peak during the SARS-CoV-2 pandemic in the early 2020s, with several countries setting up their own testing networks (including the four nations of the UK).

Next-generation sequencing in WBE

Quantification of viral burdens is not the only capability of WBE. DNA sequencing can be used to obtain information about what viral strains are found in wastewater, which can serve to detect cryptic (hidden) variants, follow trends in circulating variants, track their spread, etc. This ties in well with the ARTIC goals of using viral genome sequencing to enhance response to disease outbreaks and epidemics! There are caveats, however, to sequencing in wastewater; viral genetic material of interest is often found at very low concentrations relative to irrelevant genetic material. This makes it difficult to obtain information from pure, untargeted metagenomic sequencing, as most of sequencing data is background noise.

Tiled amplicon sequencing can be used to address this problem. Tiled amplicon sequencing in this context consists in selectively amplifying the genome of a target virus before sequencing; by having a large amount of the target relative to the background noise, it becomes more effective. This is one of the main efforts of the Quick Lab at the University of Birmingham: to develop the primer schemes necessary to perform this technique on a range of viruses. PrimalScheme, the software used to design primer schemes, was built at the Quick lab with the aim of facilitating real-time molecular epidemiology in the context of the Zika epidemic (Quick et al., 2017). Originally primer schemes were used to amplify low-concentration viruses from clinical samples, but the same advantages apply in the context of wastewater.

However, there are concerns with this approach too. The PCR reaction used to amplify the target virus relies on primers that match sections of its genome, but there is always a possibility of these primers matching unintended bits of DNA. For example, primers can match up with other primers, which can result in large amounts of bogus DNA being generated. PrimalScheme, has in-built checks that take this into account, and avoids including primers that match themselves or each other in any given scheme. Unfortunately, primers may still match up with other background DNA in a sample, which is relatively abundant and varied in wastewater samples. If this results in non-target amplification, this can negate the advantages of tiled amplicon sequencing.

Current research at the University of Birmingham

To investigate the effectiveness of tiled amplicon sequencing in complex samples such as wastewater, we have set up an experimental sampling and testing program here at the University of Birmingham. The samples are collected from the sewers of our own campus and processed in our labs at the School of Biosciences. These are weekly “grab” samples, collected simply by dipping a bucket in the wastewater stream. Though the means of sampling are rudimentary they are straightforward, and our initial tests demonstrate the presence of the faecal indicator virus crAssphage (ie., there is human waste in our water, and the extraction methods work). Additionally, we have been able to detect traces of SARS-CoV-2 and norovirus, which are a reminder of why WBE can be useful in the first place.

These samples will serve to investigate the relationship between the concentration of a virus in wastewater, and the performance of tiled amplicon sequencing in several parameters. For example, how does virus concentration relate to genome coverage and completeness? What are the most common unintended by-products of our assays in real wastewater samples? What changes in the amplicon schemes or reactions can reduce these unintended by-products? And so on. By answering these questions we can understand when it is appropriate to use tiled amplicon sequencing, how reliable results will be depending on virus concentration, and produce assays that are more specific and efficient. Ultimately, we expect the research underpinned by our wastewater collection program to improve how reliable and informative viral genome sequencing is in the context of WBE. Better WBE, in turn, can inform authorities in their response to outbreaks or in their management common diseases, helping shape policy and healthcare resource allocation, with more early warning and more efficiency than relying solely on clinical testing.

References

Karthikeyan, S., Levy, J. I., De Hoff, P., Humphrey, G., Birmingham, A., Jepsen, K., … & Knight, R. (2022). Wastewater sequencing reveals early cryptic SARS-CoV-2 variant transmission. Nature, 609(7925), 101-108.

Nemudryi, A., Nemudraia, A., Wiegand, T., Surya, K., Buyukyoruk, M., Cicha, C., … & Wiedenheft, B. (2020). Temporal detection and phylogenetic assessment of SARS-CoV-2 in municipal wastewater. Cell Reports Medicine, 1(6).

Quick, J., Grubaugh, N. D., Pullan, S. T., Claro, I. M., Smith, A. D., Gangavarapu, K., … & Loman, N. J. (2017). Multiplex PCR method for MinION and Illumina sequencing of Zika and other virus genomes directly from clinical samples. Nature protocols, 12(6), 1261-1276.

Measles sequencing update

Mon, 11 Aug 2025 00:00:00 +0000

Summary

We have recently developed the artic-measles/400/v1.0.0 primer scheme (labs). This scheme has been designed to account for all genetic variation present across all Measles virus (MeV) serotypes, allowing it to be used for samples of unknown (clinical samples) or mixed (environmental surveillance) serotypes.

This scheme also utilises 400bp amplicons, providing very high sensitivity, enabling a range of use cases, while also using the familiar ARTIC LoCost native barcoding library preparation used for SARS-CoV-2.

Design of the scheme

For in-depth details about how the scheme was designed, please see the preprint. In summary, we used the latest version of primalscheme to produce multiple primers for each primer site to account for the diversity present in genomes of all serotypes.

Validation

We performed initial validation in-house by testing the scheme against standards for 4 serotypes (A, B3, D4 and D5) and produced full coverage data with 100% base accuracy. Other collaborators have used the scheme with effective results (further communications incoming).

Opportunity to Join Pathoplexus, Loculus, & ARTIC 2

Tue, 29 Jul 2025 00:00:00 +0000

Are you excited by the idea of building global infrastructure to make pathogen sequencing more accessible, interpretable, and equitable? The Epidemiology and Virus Evolution (EVE) Group at the Swiss Tropical and Public Health Institute (Swiss TPH) is offering a unique opportunity to work at the intersection of world-leading efforts:

ARTIC 2 the latest evolution of the renowned ARTIC project, enabling affordable, portable, and real-time genetic sequencing & surveillance
Pathoplexus a pioneering open-source platform redefining how viral sequence data is shared and governed globally, powered by the modular backend database software Loculus

This is a chance to be part of an international team making a very real, tangible impact on global health and data equity!

Some key aims of the position:

Design and deployment of privacy-conscious, locally-hosted genomic data infrastructures using Loculus
Extending Loculus to link seamlessly with the ARTIC 2 sequencing pipeline for local storage, as well as enabling easy ‘one-button’ upload of viral sequence data to Pathoplexus
Integration of viral and bacterial analytical pipelines into Cephalopod, the offline research engine of ARTIC 2

We’re offering 1 position in two different forms to find the perfect candidate:

A post-doctoral position for those on an academic track, with support for traditional academic outputs and career advancement: https://jobs.swisstph.ch/Vacancies/1087/Description/2
A software engineer position for those without a PhD, with support for learning about biological data & skill-building: https://jobs.swisstph.ch/Vacancies/1082/Description/2

Please see the job listings above for full descriptions and requirements! oth positions require a strong coding background (though more experience is expected for the software developer position). Please apply once for the position best suited to you.

For informal questions (not applications) please use applications[at]hodcroft.com

This job is onsite in Basel, Switzerland - 80-100% - ideally starting this year or early 2026, fixed-term for 2 years. We welcome diverse and international applicants!

Genome assemblies for a respiratory metagenomics positive control

Thu, 24 Jul 2025 00:00:00 +0000

Clinical metagenomics uses complex laboratory protocols to detect DNA and RNA from viral and bacterial pathogens. For best results, different sample types—from viscous sputum to blood with high host background—demand specific handling. For reliable routine diagnostics, these methods must detect pathogens both sensitively and precisely, demanding robust control specimens and validation procedures.

Negative controls are relatively straightforward: one can process a mock sample containing pure water or a negative sample matrix alongside real specimens, and monitor the resulting sequences for artifacts caused by contamination, barcode hopping or otherwise. Positive controls are more challenging: an ideal positive control specimen would comprise known quantities of various pathogen taxa anticipated to be found in tested specimens, as well as a high host background to mimic real samples. But culturing and maintaining such a stock of live pathogens is laborious, expensive, and unlikely to yield reproducible results after shipping and storage.

ZeptoMetrix® sells a variety of controls for diagnostic validation purposes. Their NATtrol™ Respiratory Panel 2.1 (RP2.1) reportedly contains “purified, intact bacterial cells and viral particles” that are “chemically modified to render them non-infectious and refrigerator stable”. The RP2.1 panel is marketed as containing 18 complete viral and 4 bacterial taxa including Bordetella spp., influenza viruses, coronaviruses, and adenoviruses, split into two subpanels. Albeit advertised as qualitative rather than quantitative, it is a promising commercially available control specimen for respiratory metagenomics.

ZeptoMetrix® provides a datasheet documenting the strains included in the each RP2.1 subpanel, yet does not disclose genome sequences. As part of evaluation of the RP2.1 panel by researchers from the ARTIC network at the University of Birmingham, we have released draft quality nanopore assemblies for 16/18 virus genomes contained in RP2.1 with >90% coverage. These consensus assemblies are publicly available and archived on Zenodo. We have also developed a simple workflow for rapidly estimating sequencing coverage of RP2.1—or any other references—using k-mer containment with Sourmash.

Lab methods

Typically, viral metagenomic sequencing directly from clinical samples results in poor sensitivity due to large host backgrounds and low viral abundance. To enrich for the viruses described in the Zeptometrix control, an adapted version of the Oxford Nanopore Technologies (ONT) Rapid Metagenomic Sequencing protocol following the viral sample preparation arm was used. This method is based on a SMART (Switching Mechanism at the 5′ end of RNA Template) approach and uses random priming for cDNA synthesis followed by PCR amplification using ONT rapid barcodes to amplify and barcode cDNA in a single step. The resulting libraries were sequenced on the ONT PromethION (R10.4.1) to generate sufficient data for assembling the RP2.1 panel genomes.

Bioinformatic methods

ONT PromethION reads were basecalled with model version 4.3.0 HAC. Initial reference sequences were selected using strain information provided in the panel’s datasheet. Consensus sequences were generated using Minimap2 -x map-ont and Kindel prior to polishing with Dorado.

Draft genomes: github.com/bede/zmrp

Containment estimation workflow: github.com/bede/knownknowns

Genome	Abbreviation	Reference	Type	Length	Assembled
Adenovirus Type 1	AdV-1	AC_000017.1	DNA	35,676	✅
Adenovirus Type 3	AdV-B	DQ099432.4	DNA	35,072	✅
Adenovirus Type 31	AdV-31	AM749299.1	DNA	33,755	✅
Influenza A H1N1 A/NY/02/2009	Flu-A-H1N1-S	KT180555.1	RNA	13,130	✅
Influenza A H3N2 A/Brisbane/10/07	Flu-A-H3N2	KJ609211.1	RNA	13,290	✅
Influenza AH1 A/New Caledonia/20/99	Flu-A-H1N1-F	CY033629.1	RNA	13,292	✅
Influenza B B/Florida/02/06	Flu-B	CY018371.1	RNA	14,222	✅
Metapneumovirus 8 Peru6-2003	HMPV	OL794390.1	RNA	13,149	✅
Parainfluenza Type 1	HPIV-1	PV660323.1	RNA	15,412	✅
Parainfluenza Type 2	HPIV-2	AF533012.1	RNA	15,654	✅
Parainfluenza Type 3	HPIV-3	KY674922.1	RNA	15,382	✅
Parainfluenza Type 4	HPIV-4	EU627591.1	RNA	17,132	⚠️ gaps
Rhinovirus 1A	HRV-1A	KC894166.1	RNA	7,096	✅
RSV A	RSV-A	KY967364.1	RNA	14,855	✅
SARS-CoV-2 USA-WA1/2020	SARS-CoV-2	ON311149.1	RNA	29,778	✅
Coronavirus 229E	HCoV-229E	OZ035244.1	RNA	26,841	✅

New ARTIC website

Tue, 22 Jul 2025 00:00:00 +0000

The ARTIC Network project has a new website that focuses more on resources, documentation and knowledge.

Now live, the site can be found here (if you are not looking at it already):

http://artic.network

ARTIC 2 project starts

Thu, 01 May 2025 00:00:00 +0000

The official start of the ARTIC 2 grant and project.

ARTIC-1 project starts

Tue, 01 Aug 2017 00:00:00 +0000

The official start of the ARTIC 1 grant and project.