SARS-CoV-2 genomic surveillance from community-distributed rapid antigen tests ============================================================================== * Isla E. Emmen * William C. Vuyk * Andrew J. Lail * Sydney Wolf * Eli J. O’Connor * Rhea Dalvie * Maansi Bhasin * Aanya Virdi * Caroline White * Nura R. Hassan * Alex Richardson * Grace VanSleet * Andrea Weiler * Savannah Rounds-Dunn * Kenneth Van Horn * Marc Gartler * Jane Jorgenson * Michael Spelman * Sean Ottosen * Nicholas R. Minor * Nancy Wilson * Thomas C. Friedrich * David H. O’Connor ## Abstract For the first several years of the pandemic, SARS-CoV-2 genomic surveillance relied almost entirely on residual diagnostic specimens from nucleic acid amplification-based tests (NAATs). The use of NAATs waned after the end of the Stafford Act Emergency Declaration for COVID-19 in the US. To continue to monitor the spread and evolution of SARS-CoV-2, we partnered with local- and state-level public health agencies and the Dane County Public Library System to obtain SARS-CoV-2 genome sequence from freely available community rapid antigen tests (RATs). From August 15, 2023 to February 29, 2024 we received 227 tests through this system, from which we generated 127 sequences with >10x depth of coverage for ≥90% of the SARS-CoV-2 genome. Our results demonstrate that collecting and sequencing from RATs in partnership with trusted community stakeholders is a practical, useful approach for sustaining genomic surveillance for SARS-CoV-2. Keywords * SARS-CoV-2 * Public Health * Wisconsin * RNA * COVID-19 ## Introduction Genomic surveillance, or using genomic sequencing to track a pathogen’s evolution through time and space, is a powerful tool that can inform public health responses to disease outbreaks (1). During the COVID-19 pandemic, genomic surveillance data were used to identify variants of concern, investigate patterns of transmission, and develop effective vaccines. For example, routine genomic surveillance enabled early detection of the BA.2.86 SARS-CoV-2 variant of concern in Denmark in August 2023 (2). BA.2.86 and its sub-lineages, which include JN.1, make up the majority of reported SARS-CoV-2 sequences in the United States according to the CDC’s COVID Data Tracker as of this writing in July 2024 (3). Early identification and characterization allowed investigators to assess protective immunity afforded by vaccines against this new variant (4) and establish global monitoring. Genomic surveillance requires a large, representative set of samples from which to obtain sequences. Throughout the pandemic, nucleic acid amplification tests (NAATs) have been the gold standard for detecting SARS-CoV-2 infection (5). Many laboratories developed workflows to conduct genomic surveillance by sequencing residual nasal swab samples collected for NAAT testing. After the termination of the Stafford Act Emergency Declaration for COVID-19 in the US on May 11, 2023, many facilities stopped offering NAATs for routine COVID-19 diagnosis. During the year of 2020, the average number of NAATs performed each week was ∼587,975, which decreased to a weekly average of ∼96,215 tests during the year of 2023 (3). Subsequently, the major source of samples for genomic surveillance was greatly diminished. The CDC now recommends at-home COVID-19 rapid antigen tests (RATs) as an alternative for those without access to NAATs (6). RATs are cheaper than NAATs, provide results faster, and do not require trained personnel to perform them (7). RATs usually require the user to swab the insides of both nostrils, place the swab into an inactivation buffer and drop the buffer onto a lateral flow test. The inactivation buffer disrupts the virus membrane, exposing the nucleocapsid protein. As the liquid moves up the strip, immune complexes containing labeled antibodies conjugated with the nucleocapsid bind to a test line of antibodies if virus is present (8). As of this writing in July 2024, there are 38 different over-the-counter antigen test products authorized by the FDA in the US (9). RATs have high specificity, but have been critiqued for their lower sensitivity compared to NAATs – Smith-Jeffcoat et al. found that RATs had a sensitivity of 47% compared to NAATs (10). Another significant limitation of RATs is that users of these tests are not required to report results to any public health agency (11). Recognizing the gap in surveillance, multiple groups have investigated the potential for RATs to serve as a source material for SARS-CoV-2 sequencing to complement genomic surveillance efforts. Several methods have been developed to extract SARS-CoV-2 viral nucleic acid from antigen tests (12–16). These studies extracted SARS-CoV-2 viral RNA (vRNA) from different parts of a RAT: some have used the testing strip (12,15–17), some have used the swab (14,15), and others have used the inactivation buffer (18). All of these substrates contain enough genetic material to support viral sequencing to varying degrees. Many of these studies have demonstrated that it is possible to obtain genomes assigned as specific lineages (12,15,16), allowing for circulating lineages to be tracked. One study suggested that RATs are comparable to NAATs in their performance in qPCR and sequencing (19), and thus should be effective sample alternatives for genomic surveillance. Some of these studies began by using known concentrations of SARS-CoV-2 vRNA to spike a RAT and demonstrate the efficiency of recovering vRNA from these tests (12,16,17). Others collected real patient samples from hospitals (12) or from study volunteers (15) to demonstrate these methods could work in real-world conditions. One study demonstrated the real-world use of these tests for genomic surveillance by sequencing off of positive RATs collected in Sicily, Italy as a part of routine clinical practice to track the spread of variants over a period of six months (18). Demonstrating that sequencing SARS-CoV-2 genomes from positive RATs is technically possible was an important step, but for RAT-based sequencing to be practical for widespread genomic surveillance, there needs to be a comprehensive workflow for obtaining SARS-CoV-2-positive RATs from the community, since these “at-home” tests are typically discarded as soon as the results are read. We hypothesized that people would be willing to redirect SARS-CoV-2-positive RATs for genomic surveillance if the process was sufficiently simple. To test this hypothesis, we created a mechanism for individuals to anonymously submit positive RATs. We partnered with community organizations in Dane County, Wisconsin that were providing free RATs to bundle research packets along with the tests, which community members could use to submit positive tests through the United States Postal Service (USPS). We developed an efficient, robust method to extract SARS-CoV-2 RNA from RATs and generated SARS-CoV-2 sequences. ## Materials and Methods ### Ethics statement The University of Wisconsin IRB determined that this human research was exempt because participants were anonymous and self-identified. A secure database was created to store limited metadata including the date and location as a census block group for each test. The census block group is a geographical unit used by the US Census Bureau, usually containing between 250 and 550 housing units. ### Collection of Rapid Antigen Tests In Wisconsin, the Dane County Public Library System and Public Health Madison Dane County (PHMDC) distributed free RATs to the public. Tests were provided by the Wisconsin Department of Health Services, which received RATs from the US national stockpile. There are 29 libraries and branches in the Dane County Library System. We asked libraries in diverse parts of Madison and Dane County to participate in our program to collect RATs. Four of these libraries are in Madison and five are in more rural areas of Dane County (Figure 1). We also partnered with PHMDC, which distributed RATs at two walk-in sites. ![Figure 1.](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2024/08/14/2024.08.12.24311680/F1.medium.gif) [Figure 1.](http://medrxiv.org/content/early/2024/08/14/2024.08.12.24311680/F1) Figure 1. Map showing the locations of all Dane County libraries and Public Health Madison Dane County (PHMDC) clinics. The scale bar represents 10 miles. Dane County and Sauk County are labeled in the bottom left corner of the county boundaries. Nine of the Dane County libraries and two PHMDC sites were involved in handing out research packets along with SARS-CoV-2 rapid antigen tests to patrons, which allowed willing participants to send their positive tests to our laboratory to be sequenced. The nine libraries that were active in this program are shown as blue circles. The two PHMDC clinics that also distributed research packets with their rapid antigen tests are shown as red triangles. The gray circles depict the locations of other libraries in the Dane County Library System that were not involved in the program. To enable the public to send positive tests obtained from these locations to our lab, we designed a packet of materials to attach to each RAT kit (Figure 2). This packet included a bubble mailer with a business reply mail shipping label, a zip-lock bag with a unique barcode inside which was placed inside the mailer, and an instructional flyer affixed to the outside. The flyer had instructions in both English and Spanish, describing the study and providing instructions to participate. Users of USPS business reply mail are only charged for items that are returned, rather than for all self-addressed prepaid envelopes printed. This dramatically reduced the cost of providing postage prepaid envelopes for the study. ![Figure 2.](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2024/08/14/2024.08.12.24311680/F2.medium.gif) [Figure 2.](http://medrxiv.org/content/early/2024/08/14/2024.08.12.24311680/F2) Figure 2. Research packet description. An image of the packet that was attached to SARS-CoV-2 rapid antigen test boxes, enabling participants to send their positive tests to the lab through the USPS. In the upper left is a folded flyer attached to an envelope. The flyer explains the goal of our study and how to participate in both English and Spanish. The upper right shows a zip-lock bag with a QR code placed inside, which participants scanned to document the date and location of their test. The location of the scanned QR code was immediately converted to the census block group of the scan and stored in our secure database. The bottom shows a business-reply shipping label pasted onto the envelope, allowing participants to drop their sealed envelope at a post office drop box to be sent to our lab. We designed a secure website and database using Node JS to collect the date and location at the time a user submits their information to the website, assuming that this is a reasonable proxy for RAT assay date and location. When participants scanned randomly generated unique QR codes included in the research packets, the barcode associated with the QR code, the date of the scan, and the location, which was converted to a census block group, were populated into a MySQL database on a secure server. Users can translate website content into dozens of languages using an embedded Google Translate pull-down menu. Participants were then asked to place their test inside the provided zip-lock bag, seal the zip-lock bag, place it in a postage prepaid business-reply mailer, and drop the sealed envelope at any post office mailbox. The inactivation buffer in a RAT inactivates the SARS-CoV-2 virus, rendering the tests non-biohazardous and therefore safe to send through the USPS (20). When tests arrived at the lab, we scanned the QR code into our database to record the date received and then stored each RAT at -80°C until processing. The majority of RATs we received were BinaxNOWTM COVID-19 Antigen Self Tests (Abbott, [https://www.abbott.com](https://www.abbott.com)) or iHealth COVID-19 Antigen Rapid Tests (iHealth Labs Inc., [https://ihealthlabs.com](https://ihealthlabs.com)), but there were a few samples of other brands. ### Extraction of Nucleic Acids Originally, we attempted to isolate nucleic acids from RATs using the protocol described by Martin et al (12). We optimized this protocol as described below: Tests were stored at -80°C until we processed them in batches. They were thawed and opened to retrieve the testing strip, and swab, if included, with sterile disposable forceps. Instead of cutting the strip into three pieces, the whole strip was placed into a clean 5 mL Sarstedt freezer tube (Sarstedt, [https://www.sarstedt.com/en](https://www.sarstedt.com/en)). Instead of adding 800 μL of Buffer AVL from the QIAmp Viral RNA mini kit (QIAGEN, [https://www.qiagen.com/us](https://www.qiagen.com/us)) we added 800 μL of Viral Transport Medium (VTM, Rocky Mountain Biologicals, LLC, [https://rmbio.com](https://rmbio.com)) to the 5 mL tube, which was incubated at room temperature for ten minutes on a Hulamixer (ThermoFisher Scientific, [https://www.thermofisher.com](https://www.thermofisher.com)). 500 μL were transferred to a clean 1.5 mL tube, and 5 μL of Dynabeads Wastewater Virus Enrichment Beads (ThermoFisher Scientific) were added to each sample. Subsequently, we followed the manufacturer’s protocol for the MagMAX™ Wastewater Ultra Nucleic Acid Isolation Kit with Virus Enrichment (ThermoFisher Scientific). Samples were isolated on a Kingfisher Apex instrument (ThermoFisher Scientific) following the manufacturer’s protocol (MagMAX_Wastewater_DUO96.bdz). Following isolation, the samples were treated with TurboTM DNase (ThermoFisher Scientific) according to the manufacturer’s protocol. Instead of proceeding with a phenol/chloroform extraction, the samples were processed using the RNA Clean and Concentrator-5 kit (Zymo Research, [https://www.zymoresearch.com](https://www.zymoresearch.com)) following the manufacturer’s protocol and skipping the in-column DNase I Treatment. ### RT-qPCR We assumed that all RATs with a positive test line had enough vRNA to be sequenceable, so we did not test every sample with RT-qPCR. Later, we selected a random subset of 76 samples to be quantified by RT-qPCR to investigate trends between the Ct and sequencing quality. We quantified SARS-CoV-2 vRNA using the CDC N1 Taqman assay (21). To confirm successful isolation of viral nucleic acid, each sample was tested for the presence of human ribonuclease P nucleic acid. Only samples which tested positive for ribonuclease P were included in the analysis. A negative control was generated using the same VTM used on our other samples and processed following the same protocols. These were also run with the qPCR assay, and had no detectable amplification. The primers and probe for this assay are commercially available from Integrated DNA Technologies (IDT, [https://www.idtdna.com](https://www.idtdna.com)). The assay was run on a LightCycler 96 instrument (Roche, [https://www.roche.com](https://www.roche.com)) using the Taqman Fast Virus 1-step Master Mix enzyme (ThermoFisher Scientific). ### RT-PCR and Sequencing We performed amplicon-based SARS-CoV-2 sequencing on the isolated vRNA. Sample libraries were prepared using the QIAseq DIRECT SARS-CoV-2 Kit with Booster and Enhancer (QIAGEN) according to the manufacturer’s instructions. Samples were normalized to 4 nM and pooled together. The pool was diluted to a final concentration of 8 pM, which was then run using 2x150 Miseq Reagent Kits v2 (Illumina, [https://www.illumina.com](https://www.illumina.com)) on a MiSeq instrument (Illumina). ### Sequencing Analysis Raw sequencing reads were quality-checked, aligned to the Wuhan-1 SARS-CoV-2 reference, and variant-called, using the open-source *viralrecon* pipeline from the *nf-core* project (22–24). Details for how we ran *viralrecon*, alongside the custom R scripts we used to generate figures, are available in our Library-Rapid-Antigen-Test-Manuscript Github repository. ### Statistical Analysis The status of samples (pass or fail) was compared to both Ct and the length of transit time, using an unpaired two-tailed t-test (Prism v10.1.0). The significance cutoff was p < 0.05. The mean and standard deviation of the number of tests received each month was calculated using Microsoft Excel v16.78. We compared our data to the Wisconsin State Lab of Hygiene’s (WSLH) SARS-CoV-2 Wastewater Genomic Dashboard (25) by recording the two most prevalent lineage groups detected by WSLH in the state of Wisconsin for each two week period between August 28th, 2023 to February 25th, 2024. We separated our passing RAT sequences into two week periods by their test scan date with the assumption that the barcode was scanned on the date the test was taken. We determined Pango lineages for our RATs by running the consensus sequences through Nextclade v3.5.0 (26). We compared the lineages detected through our RAT program to the two most prevalent lineage groups reported by WSLH to determine how frequently our program detected the two most prevalent lineage groups. ## Results ### Test Collection Between August 15th, 2023, and February 29th, 2024, we supplied nine Dane County libraries and two public health clinics with approximately 7,775 research packets to attach to the RATs distributed to patrons. We received 223 packets, which amounts to a return rate of ∼2.9%. Some of these packets contained multiple tests, resulting in a total of 227 tests. The return rates varied by month (Table 1), but the mean number received each month was 32 ± 10. View this table: [Table 1.](http://medrxiv.org/content/early/2024/08/14/2024.08.12.24311680/T1) Table 1. Monthly rate of materials supplied to collect positive rapid antigen tests, positive tests received in our lab, and tests that passed our sequencing quality threshold Some tests came in without their barcode, causing us to lose the associated metadata. There were also tests that arrived with a barcode which had not been scanned by the participant. Of the 223 research packets received, 170 were properly associated with time and location metadata. For those that did not have metadata, we used the date received as a proxy for the time of the test, but we were unable to assign a location to the test. Of the received tests for which we have location data, one was scanned in Sauk County, Wisconsin and the rest were scanned in Dane County, Wisconsin (Figure 1). ### Sequencing Quality We attempted to sequence SARS-CoV-2 from all 227 RATs. We considered a sequence with genome coverage ≥ 90% at a depth of coverage > 10x to be a “passing” sequence. 127 of the 227 RAT-derived sequences passed: a success rate of 56%. Of the 76 samples we tested with RT-qPCR, 15 had no detectable amplification of the N1 target sequence. We obtained passing sequences for samples with Cts up to 35.4. The mean Ct for samples that passed was 31.8 and the mean Ct for those that failed was 35.3 (Figure 3A). There was a significant difference in Ct between these two groups (unpaired t-test, two-tailed p < 0.0001, df = 59). ![Figure 3.](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2024/08/14/2024.08.12.24311680/F3.medium.gif) [Figure 3.](http://medrxiv.org/content/early/2024/08/14/2024.08.12.24311680/F3) Figure 3. Amount of viral material correlated with our ability to sequence rapid antigen test (RAT) samples, but time en route did not. A) A scatterplot comparing the Cycle Threshold (Ct) value obtained through quantitative PCR (qPCR) to the percent coverage of the SARS-CoV-2 genome at >10x depth of coverage obtained from each RAT sample. For a sequence to pass the sequencing quality threshold, a sequence must cover ≥90% of the SARS-CoV-2 genome at >10x depth. Samples that passed this threshold are colored blue, and those that did not are colored red. The mean Ct for samples that passed was 31.78 and the mean Ct for those that failed was 35.34 (unpaired t-test, two-tailed p < 0.0001, df = 59). Samples with lower Cts were correlated with higher SARS-CoV-2 coverage. B) This scatterplot shows the time (in days) between when the QR code was scanned by a participant and when our lab received the sample, referred to as the transit time. These tests are separated into those that passed the sequencing quality threshold (blue circles) as defined above and those that failed (red triangles). The horizontal black line is the median value for each group. The mean transit time for the samples that passed our sequencing quality metrics was 6.6 days and the mean transit time for samples that failed was 6.3 days. There was no significant difference between these groups (unpaired t-test, p = 0.58). There was variability in the transit time, i.e., the amount of time between when the tests were scanned by the participant and when we received them in the lab, ranging from one day to twenty days. We investigated whether transit time affected whether a sequence passed. Only samples with a participant-scanned date were included in this analysis. Figure 3B illustrates the spread of transit times compared to the tests that passed or failed. The mean transit time for the samples that passed was 6.6 days, compared to 6.3 days for samples that failed. These two means are not statistically different (unpaired two tailed t-test, p = 0.58). These results suggest that the amount of time samples sat at uncontrolled temperatures did not have a significant impact on RNA sequencing quality within the range of the time in transit investigated in this study. ### Tracking SARS-CoV-2 Lineages We used Nextclade v3.5.0 (26) to obtain the Pango lineage of each sample that passed. The variation in frequencies of the SARS-CoV-2 lineages of our samples over each week can be seen in Figure 4. From August to November 2023, the majority of lineages were assigned to the XBB.1* parent and XBB.2* parent lineages. Beginning in December 2023, we observed a shift to the lineage JN.1*, which made up the majority of our RAT sequences in January and February 2024. ![Figure 4.](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2024/08/14/2024.08.12.24311680/F4.medium.gif) [Figure 4.](http://medrxiv.org/content/early/2024/08/14/2024.08.12.24311680/F4) Figure 4. Rapid antigen test lineages determined for each week. This chart shows the proportion of SARS-CoV-2 lineages determined for passing (≥90% of the SARS-CoV-2 genome at >10x depth) rapid antigen test (RAT) samples by week. The date used was the date the QR code was scanned by the participant. RATs that were not scanned were excluded from this analysis. From August to mid-November, the most common lineages in our samples fell under XBB.1.5*, XBB.1.9.2*, XBB.1.16*, and XBB.2.3. Beginning in early December, we began to see an increase in the number of samples belonging to the lineage JN.1*. This lineage dominated RAT samples scanned in February 2024. We compared the lineages observed in our passing RAT sequences to the WSLH’s SARS-CoV-2 Wastewater Genomic Dashboard (25) to see how well our program picked up the most frequently detected lineages in wastewater around the state of Wisconsin (Table 2). During 12 of the 13 two-week reporting periods on WSLH’s Genomic Dashboard we detected the dominant circulating lineage in our RATs. We detected the second most prevalent lineage in the WSLH’s Genomic Dashboard during 7 of the 13 reporting periods. The concordance between our results and WSLH’s wastewater program gave us confidence that our sequences accurately reflected circulating lineages in the community. Our program was also able to detect lineages earlier than other existing genomic surveillance programs that rely on individual tests in the state of Wisconsin. Our RAT sequences account for six of the earliest documented cases of a lineage in Wisconsin in Genbank and GISAID as of April 18, 2024: JN.1.1, JN.1.2, XDD, XDA, XDP, and XDE (Table 3), highlighting the ability of this program to fill in gaps of detection and to augment other methods of genomic surveillance. View this table: [Table 2.](http://medrxiv.org/content/early/2024/08/14/2024.08.12.24311680/T2) Table 2. List of lineages of our rapid antigen test passing sequences that corresponded to the two most prevalent lineage groups in the wastewater signal in the state of Wisconsin for each two week period reported by the WSLH View this table: [Table 3.](http://medrxiv.org/content/early/2024/08/14/2024.08.12.24311680/T3) Table 3. Rapid antigen test samples that are the earliest recorded example of their respective Pango lineage in the state of Wisconsin according to data submitted to GISAID and Genbank as of April 18, 2024 ## Discussion Genomic surveillance has been very effective in tracking SARS-CoV-2 evolution throughout the COVID-19 pandemic, guiding vaccine formulation and identifying variants of concern early on (27). Sequencing from residual NAAT nasal swabs was the basis of a robust national genomic surveillance program. Currently most individuals use RATs to test for COVID-19 instead of NAATs. Testing results of RATs are usually not reported to public health organizations and most tests are discarded, restricting their use as a source of samples for genomic surveillance. We saw an opportunity to evaluate the feasibility of using RATs for genomic surveillance by creating a system for their collection and developing a method of sequencing the nucleic acids present. Our findings demonstrate that collecting and sequencing SARS-CoV-2 from RATs is a tractable way to conduct genomic surveillance. Data collected by PHMDC from August 12, 2023 to February 24, 2024 from a subset of labs conducting NAATs suggests that the percent test positivity for COVID-19 in Dane County averaged 12.3% during this period and ranged from 7.8% to 16% (28). If we assume that average test positivity during our program was 12.3% in Dane County, we received about a quarter of all positive tests from patrons who had received our research packets, given our return rate of the packets was 2.9%. We believe this is a reasonable return rate because we are relying on sick individuals to voluntarily mail their tests to us without any direct benefit to themselves. The method of sequencing we developed yielded strong coverage of SARS-CoV-2 extracted from RATs. We obtained sequences that passed our quality threshold for samples with Cts up to 35.4. These results support the findings of other studies (12,15,19), which show that sequencing from RATs can be done reliably. Our study demonstrated that sequencing is possible from real-world RATs collected voluntarily from people in surrounding communities. The transit time during which RATs sat in uncontrolled conditions had a negligible impact on overall sequencing success, suggesting that transportation through the mail is a viable method for the collection of RATs, even in Wisconsin’s challenging winter climate. Other studies have demonstrated that extraction of nucleic acids is possible from RATs stored at room temperature (RT) for long periods of time (13,15); one study generated 75.2% genome coverage from a RAT stored at RT for 3 months. We obtained a sequence with >10x coverage for ≥90% of the SARS-CoV-2 genome from a RAT that sat at uncontrolled temperatures for at least 17 days. Our RAT genomic surveillance program showed good concordance with the WSLH’s SARS-CoV-2 Wastewater Genomic Dashboard (25), which is based on sequencing results from wastewater collected around the state of Wisconsin, giving us confidence in the accuracy of our surveillance sequencing. Moreover, our program found early examples of newly emerging lineages in our community, suggesting that this program filled in gaps of surveillance in the community. We found both nascent and rare lineages, such as JN.1 and XDE, respectively. The latter has only been documented 22 times in North America according to CoV-Spectrum (29). Therefore, we believe that our program for collecting and sequencing from positive RATs is an important addition to wastewater and NAAT-based approaches to genomic surveillance. An important limitation of our study is the reliance on self-reported data, which is less precise than clinical specimen metadata. The metadata depends on the participant’s QR code scan to approximate the date and location of the test, which was used instead of the actual test date in order to maximize participant privacy by minimizing identifiability. Participants might scan their QR code days or weeks after taking the test, reducing date accuracy, though we speculate that this is unlikely. Another limitation is that 54 of the 223 envelopes arrived unscanned, resulting in no associated meta-data. These challenges are inherent when working with community participants unfamiliar with the study protocol. We updated the instructions provided to participants to address issues as they arose, but it is not possible to ensure all participants read the instructions carefully. Our program could be scalable given the right conditions: sufficient funding, strong community engagement and trust, and a source of RATs to attach to the packets, as uptake might be slower without the incentive of free RATs. With a return rate of 2.9%, our material cost per successfully returned RAT packet was about $36, not including sequencing costs, which could scale well with proper funding. In designing our study, we leveraged an existing, robust system for providing free RATs to the community, which allowed us to efficiently implement this program. Building off of similar existing systems could allow for efficient replication of this program and may help to increase community engagement and trust. The creation of packets is a time-consuming process, which could limit the overall scalability of this process. A potential drawback of a program like this for public health purposes is the slower turnaround for sequencing results. The time each of our RATs spent in transit varied greatly, with samples being received up to 20 days after participant scan, meaning that the results may not be as actionable for public health. As RATs become available for other viruses, this program could also be expanded to include genomic surveillance for these pathogens. The FDA has authorized at-home antigen tests for Influenza A and RSV in the US, and these tests could also be collected to conduct genomic surveillance on these viruses (30–32). Some studies have demonstrated the possibility of recovering various other respiratory viruses from COVID-19 RATs, with varying levels of success across different viral taxa (10,14). Additionally, collecting both positive and negative RATs could help estimate the prevalence of respiratory viruses present in the community, thereby increasing the program’s public health relevance. Continuing to conduct genomic surveillance in a post-pandemic world is challenging. Fewer people are getting tested for COVID-19, even when experiencing COVID-like symptoms, making it increasingly difficult to collect positive tests. Most individuals now use RATs for COVID-19 testing, making them an attractive sample source, but few systems exist to collect them from the community. We designed a system for the collection and sequencing of positive RATs that can effectively augment current genomic surveillance programs and inform public health about circulating SARS-CoV-2 strains in a community. ## Supporting information Supplementary Table 1 [[supplements/311680_file02.pdf]](pending:yes) IRB Determination [[supplements/311680_file03.pdf]](pending:yes) ## Data Availability The sequencing data generated in this study are available in the Sequence Read Archive (SRA) under the BioProject PRJNA1096364. The accession numbers to the sequences used in these analyses are available in Supplementary Table 1. ## Acknowledgments This work was supported by the Wisconsin Department of Health Services [435100-A24-ELCProjE] and the CDC [75D30122C15355]. Analysis of these data was made possible by the Center for High Throughput Computing’s High Performance Cluster at the University of Wisconsin-Madison. The Wisconsin Department of Health Services provided RATs to libraries and the public, during and after the COVID-19 public emergency. We would like to thank the many patrons of Dane County Public Libraries and PHMDC who were kind enough to take time while they were sick to scan and send us their tests. Without these community members, this study would not be possible. We would also like to thank all the staff at the Dane County Public Libraries who helped us to implement this program. We would like to specifically thank Leah Fritsche, Erick Plumb, Matt Rahner, Samantha Seeman, and Elizabeth Clauss, who agreed to distribute these packets at their libraries. We would also like to thank Sam Petykowski, a high school student who volunteered his time to produce research packets for Alicia Ashman Library. * Received August 12, 2024. * Revision received August 12, 2024. * Accepted August 14, 2024. * © 2024, Posted by Cold Spring Harbor Laboratory This pre-print is available under a Creative Commons License (Attribution-NonCommercial 4.0 International), CC BY-NC 4.0, as described at [http://creativecommons.org/licenses/by-nc/4.0/](http://creativecommons.org/licenses/by-nc/4.0/) ## References 1. 1.Ladner JT, Sahl JW. Towards a post-pandemic future for global pathogen genome sequencing. PLOS Biol. 2023 Aug 1;21(8):e3002225. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1371/journal.pbio.3002225&link_type=DOI) 2. 2.Rasmussen M, Møller FT, Gunalan V, Baig S, Bennedbæk M, Christiansen LE, et al. First cases of SARS-CoV-2 BA.2.86 in Denmark, 2023. Eurosurveillance [Internet]. 2023 Sep 7 [cited 2024 Apr 30];28(36). Available from: [https://www.eurosurveillance.org/content/10.2807/1560-7917](https://www.eurosurveillance.org/content/10.2807/1560-7917). ES.2023.28.36.2300460 3. 3.CDC. Centers for Disease Control and Prevention. 2020 [cited 2024 May 9]. COVID Data Tracker. Available from: [https://covid.cdc.gov/covid-data-tracker](https://covid.cdc.gov/covid-data-tracker) 4. 4.Ma KC, Shirk P, Lambrou AS, Hassell N, Zheng X yu, Payne AB, et al. Genomic Surveillance for SARS-CoV-2 Variants: Circulation of Omicron Lineages — United States, January 2022–May 2023. MMWR Morb Mortal Wkly Rep. 2023 Jun 16;72(24):651–6. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.15585/mmwr.mm7224a2&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=37319011&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F08%2F14%2F2024.08.12.24311680.atom) 5. 5.World Health Organization. Recommendations for national SARS-CoV-2 testing strategies and diagnostic capacities. 6. 6.CDC. Centers for Disease Control and Prevention. 2020 [cited 2024 Apr 30]. COVID-19 and Your Health. Available from: [https://www.cdc.gov/coronavirus/2019-ncov/symptoms-testing/testing.html](https://www.cdc.gov/coronavirus/2019-ncov/symptoms-testing/testing.html) 7. 7.Khalid MF, Selvam K, Jeffry AJN, Salmi MF, Najib MA, Norhayati MN, et al. Performance of Rapid Antigen Tests for COVID-19 Diagnosis: A Systematic Review and Meta-Analysis. Diagnostics. 2022 Jan 4;12(1):110. 8. 8.ASM.org [Internet]. [cited 2024 Jul 25]. How the SARS-CoV-2 EUA Antigen Tests Work. Available from: [https://asm.org:443/Articles/2020/August/How-the-SARS-CoV-2-EUA-Antigen-Tests-Work](https://asm.org:443/Articles/2020/August/How-the-SARS-CoV-2-EUA-Antigen-Tests-Work) 9. 9.Health C for D and R. At-Home OTC COVID-19 Diagnostic Tests. FDA [Internet]. 2024 Jul 9 [cited 2024 Jul 16]; Available from: [https://www.fda.gov/medical-devices/coronavirus-covid-19-and-medical-devices/home-otc-covid-19-diagnostic-tests](https://www.fda.gov/medical-devices/coronavirus-covid-19-and-medical-devices/home-otc-covid-19-diagnostic-tests) 10. 10.Smith-Jeffcoat SE, Mellis AM, Grijalva CG, Talbot HK, Schmitz J, Lutrick K, et al. SARS-CoV-2 Viral Shedding and Rapid Antigen Test Performance — Respiratory Virus Transmission Network, November 2022–May 2023. MMWR Morb Mortal Wkly Rep. 2024 Apr 25;73(16):365–71. 11. 11.Ritchey MD, Rosenblum HG, Del Guercio K, Humbard M, Santos S, Hall J, et al. COVID-19 Self-Test Data: Challenges and Opportunities — United States, October 31, 2021–June 11, 2022. MMWR Morb Mortal Wkly Rep. 2022 Aug 12;71(32):1005–10. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.15585/mmwr.mm7132a1&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=35951486&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F08%2F14%2F2024.08.12.24311680.atom) 12. 12.Martin GE, Taiaroa G, Taouk ML, Savic I, O’Keefe J, Quach R, et al. Maintaining genomic surveillance using whole-genome sequencing of SARS-CoV-2 from rapid antigen test devices. Lancet Infect Dis. 2022 Oct;22(10):1417–8. 13. 13.Rector A, Bloemen M, Schiettekatte G, Maes P, Van Ranst M, Wollants E. Sequencing directly from antigen-detection rapid diagnostic tests in Belgium, 2022: a gamechanger in genomic surveillance? Eurosurveillance [Internet]. 2023 Mar 2 [cited 2024 Apr 30];28(9). Available from: [https://www.eurosurveillance.org/content/10.2807/1560-7917.ES.2023.28.9.2200618](https://www.eurosurveillance.org/content/10.2807/1560-7917.ES.2023.28.9.2200618) 14. 14.Paull JS, Petros BA, Brock-Fisher TM, Jalbert SA, Selser VM, Messer KS, et al. Optimisation and evaluation of viral genomic sequencing of SARS-CoV-2 rapid diagnostic tests: a laboratory and cohort-based study. Lancet Microbe. 2024 Apr;S2666524723003993. 15. 15.Nguyen PV, Carmola LR, Wang E, Bassit L, Rao A, Greenleaf M, et al. SARS-CoV-2 molecular testing and whole genome sequencing following RNA recovery from used BinaxNOW COVID-19 antigen self tests. J Clin Virol. 2023 May;162:105426. 16. 16.Macori G, Russell T, Barry G, McCarthy SC, Koolman L, Wall P, et al. Inactivation and Recovery of High Quality RNA From Positive SARS-CoV-2 Rapid Antigen Tests Suitable for Whole Virus Genome Sequencing. Front Public Health. 2022 May 3;10:863862. 17. 17.Moso MA, Taiaroa G, Steinig E, Zhanduisenov M, Butel-Simoes G, Savic I, et al. Non-SARS-CoV-2 respiratory viral detection and whole genome sequencing from COVID-19 rapid antigen test devices: a laboratory evaluation study. Lancet Microbe. 2024 Apr;5(4):e317–25. 18. 18.Cucina A, Contino F, Brunacci G, Orlando V, La Rocca M, Indelicato S, et al. The Suitability of RNA from Positive SARS-CoV-2 Rapid Antigen Tests for Whole Virus Genome Sequencing and Variant Identification to Maintain Genomic Surveillance. Diagn Basel Switz. 2023 Dec 7;13(24):3618. 19. 19.Hassouneh SAD, Trujillo A, Ali S, Cella E, Johnston C, DeRuff KC, et al. Antigen test swabs are comparable to nasopharyngeal swabs for sequencing of SARS-CoV-2. Sci Rep. 2023 Jul 12;13(1):11255. 20. 20.Coelho FF, Da Silva MA, Lopes TB, Polatto JM, De Castro NS, Andrade LAF, et al. SARS-CoV-2 Rapid Antigen Test Based on a New Anti-Nucleocapsid Protein Monoclonal Antibody: Development and Real-Time Validation. Microorganisms. 2023 Sep 28;11(10):2422. 21. 21.Lu X, Wang L, Sakthivel SK, Whitaker B, Murray J, Kamili S, et al. US CDC Real-Time Reverse Transcription PCR Panel for Detection of Severe Acute Respiratory Syndrome Coronavirus 2. Emerg Infect Dis. 2020 Aug;26(8):1654–65. [PubMed](http://medrxiv.org/lookup/external-ref?access_num=http://www.n&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F08%2F14%2F2024.08.12.24311680.atom) 22. 22.Langer BE, Amaral A, Baudement MO, Bonath F, Charles M, Chitneedi PK, et al. Empowering bioinformatics communities with Nextflow and nf-core [Internet]. 2024 [cited 2024 May 29]. Available from: [http://biorxiv.org/lookup/doi/10.1101/2024.05.10.592912](http://biorxiv.org/lookup/doi/10.1101/2024.05.10.592912) 23. 23. Harshil Patel, Monzón S, Varona S, Espinosa-Carrasco J, Garcia MU, Nf-Core Bot, et al. nf-core/viralrecon: nf-core/viralrecon v2.6.0 - Rhodium Raccoon [Internet]. Zenodo; 2023 [cited 2024 May 29]. Available from: [https://zenodo.org/record/7764938](https://zenodo.org/record/7764938) 24. 24.Ewels PA, Peltzer A, Fillinger S, Patel H, Alneberg J, Wilm A, et al. The nf-core framework for community-curated bioinformatics pipelines. Nat Biotechnol. 2020 Mar;38(3):276–8. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/s41587-020-0439-x&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=32055031&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2024%2F08%2F14%2F2024.08.12.24311680.atom) 25. 25.SARS-CoV-2 Wastewater Genomic Dashboard [Internet]. [cited 2024 Jun 7]. Available from: [https://dataportal.slh.wisc.edu/sc2-ww-dashboard](https://dataportal.slh.wisc.edu/sc2-ww-dashboard) 26. 26.Aksamentov I, Roemer C, Hodcroft E, Neher R. Nextclade: clade assignment, mutation calling and quality control for viral genomes. J Open Source Softw. 2021 Nov 30;6(67):3773. 27. 27.Tosta S, Moreno K, Schuab G, Fonseca V, Segovia FMC, Kashima S, et al. Global SARS-CoV-2 genomic surveillance: What we have learned (so far). Infect Genet Evol. 2023 Mar;108:105405. 28. 28.Respiratory Illness Dashboard | Public Health Madison & Dane County [Internet]. 2023 [cited 2024 Jun 7]. Available from: [https://publichealthmdc.com/health-services/respiratory-illness/dash-board](https://publichealthmdc.com/health-services/respiratory-illness/dash-board) 29. 29.Chen C, Nadeau S, Yared M, Voinov P, Xie N, Roemer C, et al. CoV-Spectrum: analysis of globally shared SARS-CoV-2 data to identify and characterize new variants. Alkan C, editor. Bioinformatics. 2022 Mar 4;38(6):1735–7. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1093/bioinformatics/btab856&link_type=DOI) 30. 30.Health C for D and R. Influenza Diagnostic Tests. FDA [Internet]. 2024 Jul 22 [cited 2024 Aug 2]; Available from: [https://www.fda.gov/medical-devices/in-vitro-diagnostics/influenza-diagnostic-tests](https://www.fda.gov/medical-devices/in-vitro-diagnostics/influenza-diagnostic-tests) 31. 31.QuickVue® RSV Test [Internet]. [cited 2024 Aug 2]. Available from: [https://www.quidelortho.com/global/en/products/quickvue-rapid-lateral-flow-tests/quickvue-rsv-test](https://www.quidelortho.com/global/en/products/quickvue-rapid-lateral-flow-tests/quickvue-rsv-test) 32. 32.BinaxNOW RSV Card [Internet]. [cited 2024 Aug 2]. Available from: [https://www.globalpointof-care.abbott/ww/en/product-details/binaxnow-rsv.html](https://www.globalpointof-care.abbott/ww/en/product-details/binaxnow-rsv.html)