Identification of SARS-CoV-2 Variants Using Viral Sequencing for the Centers for Disease Control and Prevention Genomic U.S. Surveillance Program.

doi:10.21203/rs.3.rs-1134436/v1

Download PDF

Research Article

Identification of SARS-CoV-2 Variants Using Viral Sequencing for the Centers for Disease Control and Prevention Genomic U.S. Surveillance Program.

https://doi.org/10.21203/rs.3.rs-1134436/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

The Centers for Disease Control and Prevention contracted with laboratories to sequence the SARS-CoV-2 genome from positive samples across the United States to enable public health officials to investigate the impact of variants on disease severity as well as the effectiveness of vaccines and treatment. Herein we present the initial results from full SARS- CoV-2 viral genomic sequencing of 24,441 positive patient samples between April and June 2021.

Methods

RT-PCR confirmed (N Gene Ct value <30) positive patient samples, with nucleic acid extracted from saliva, nasopharyngeal and oropharyngeal swabs were selected for viral whole genome SARS-CoV-2 sequencing. Sequencing was performed using Illumina COVIDSeqtm protocol on either the NextSeq550 or NovaSeq6000 systems. Informatic variant calling, and lineage analysis were performed using DRAGEN COVID Lineage applications on Illumina’s Basespace cloud analytical system. All sequence data and variant calls were uploaded to NCBI and GISAID.

Results

An association was observed between higher sequencing coverage, quality, and samples with a lower Ct value, with <27 being optimal, across both sequencing platforms and sample collection methods. Both nasopharyngeal swabs and saliva samples were found to be optimal samples of choice for SARS-CoV2 surveillance sequencing studies, both in terms of strain identification and sequencing depth of coverage, with NovaSeq 6000 providing higher coverage than the NextSeq 550. The most frequent variants identified were the B.1.617.2 Delta (India) and P.1 Gamma (Brazil) variants.

Conclusion

These initial analyses highlight the importance of sequencing platform, sample collection methods, and RT-PCR Ct values in guiding surveillance efforts to inform the public of variant transmission rates, with the potential to facilitate correlation of sequencing data with vaccine efficacy rate and efficacy of infection prevention precautions.

COVID-19

SARS-CoV-2

Centers for Disease Control

Next Generation Sequencing

Reverse Transcription Polymerase Chain Reaction

Cycle Threshold

Lineage

Variant

As the number of patients confirmed to be infected with the SARS-CoV-2 virus and developing COVID-19 increased, international sequencing efforts began to determine genetic variations in SARS-CoV2 genes that may potentially increase viral transmission and pathogenicity 1. However, the number of genomes sequenced within the United States (U.S.) deposited in the online genome

repository, GISAID, in March of 2021 represented only 1.6% of the number of COVID-19 cases that month. As international sequencing efforts increased, variants of high concern were identified in the United Kingdom (SARS-CoV-2 B.1.1.7), and in the United States (B.1.1.7, B.1.351, P.1, B.1.427, and B.1.429). Among the SARS-CoV-2 variants of concern, Beta B.1.351 (South Africa) and Gamma P.1 (Brazil) have high potential to reduce the efficacy of some vaccines 2-5. The B.1.617 variant (India) has become a variant of interest for its high transmission rate and ability to evade immune responses 6. At the time of submission of this manuscript the first cases of the South African variant B.1.1.529 (Omicron) have been detected, but not yet identified in the U.S. 21 These genomic sequencing efforts have allowed scientists to identify not only SARS-CoV-2 positive patients and viral sequence variants, but to also monitor how new viral variants evolve and to understand how these changes affect the characteristics of the virus, with this information ultimately being used to better understand health impacts. In addition, these variants of concern are actively being monitored to determine if they may reduce the efficacy of currently approved vaccines against SARS-CoV-2 7.

In March of 2021, the Centers for Disease Control (CDC) contracted with large commercial diagnostic labs to sequence patient samples across the U.S. 8. These laboratories committed to sequencing up to 6,000 samples per week, with the capacity to scale up in response to the nation’s needs (Figure 1). The purpose of this program is to perform routine analysis of genetic sequence data to enable the CDC and its public health partners to identify and characterize variant viruses—either new ones identified in the U.S. or those already identified abroad—and to investigate how variants impact COVID-19 disease severity and the effectiveness of vaccines and treatment. Infinity Biologix LLC (IBX), located in Piscataway NJ, was awarded one of these contracts to participate in this surveillance program.

In this study we describe the results of full-length genomic sequencing and surveillance of the SARS-CoV-2 virus from the first 24,441 confirmed positive saliva (SA) samples and a small subset of nasopharyngeal and oropharyngeal (NP, OP) swab-based samples sequenced between April and June 2021.

Program Selection

IBX was designated by the CDC to participate in its Genomic Surveillance for SARS- CoV-2 Variants program 1. This program conducts genomic surveillance using a random sampling of previously confirmed positive samples from across the United States, including all 50 states, Washington DC, Puerto Rico and major U.S. territories and possessions. A target of up to 6,000 genomic sequences per week was established including State and Zip Code as the required demographics data. CDC requested, when possible and if available, additional demographic data as age, sex, and ethnicity of the patients.

Patients and sample acquisition

Diagnostic patient samples for SARS-CoV-2 testing arrived via multiple external test clients, approved by IBX and FDA. Demographic, vaccination, ethnicity and age were available if provided by the patient during sample collection. Most samples (>95%) were saliva-based and collected in Minnesota (MN) and New Jersey (NJ) given the presence of IBX laboratories in each state.

Criteria for SARS-CoV-2 Viral Sequencing

The criteria for positive samples selection were established using a N Gene (nucleocapsid protein) Ct value threshold of ≤ 30.0 on the Infinity BiologiX TaqPath SARS-CoV-

2 QPCR Assay. Per the CDC requirement, samples to be sequenced must have been no more than 10 days old following confirmation of a positive test result.

RNA isolation

All samples selected for sequencing had nucleic acid freshly extracted from the primary sample source independently of the material extracted for the initial RT-PCR testing. Automated RNA extraction from either SA, NP or OP was carried out using the Chemagic Viral DNA/RNA 300 Kit H96 (CMG-1033-S, PerkinElmer) on the Chemagic 360 Nucleic Acid Extractor (2024- 0020, Perkin Elmer).

Real-Time PCR for SARS-CoV-2

All samples were tested using the Infinity BiologiX TaqPath SARS-CoV-2 QPCR Assay according to the procedure detailed in FDA Emergency Use Authorization (EUA) #200090 to provide diagnostic results for the presence of SARS-CoV-2. This assay is approved for SA, OP and NP swab types 9. Briefly, this assay quantitates expression of three viral loci, the N, S and ORF1ab genes, with the N gene Ct value being germane to the selection of samples to be sequenced in this study. Samples testing positive for the presence of SARS-CoV-2 were isolated and stored at 4°C within 72 hours after collection for potential further use. Samples were transferred to -80°C for permanent storage.

Library Preparation and Sequencing

Sequencing libraries (500 initial samples tested) were prepared manually using the Illumina COVIDSeq Test protocol following the manufacturer’s guidelines (Illumina Inc., San Diego, CA.) 10,11. All remaining libraries were then prepared in 96-well PCR plates (2 plates per run) on Sciclone® G3 NGSx Workstations (Perkin Elmer Inc, Waltham, MA). Following cDNA synthesis and amplification, PCR amplicons were tagmented (fragmented and tagged with adaptors) using IDT® for Illumina PCR Unique Dual Indexes Set 1-4 (384 Indexes). Libraries were purifiedand pooled following the manufacture’s guidelines (Illumina Inc., San Diego, CA.). Library pools (each containing 94 samples, 1 CPC HT, 1 NTC) were quantified and analyzed for fragment size on a Tapestation 4200 (Agilent Technologies, Inc., Santa Clara, CA). Pools (96 samples) were then normalized to 4nM and combined into a sequencing pool for a total of 384 samples (index sets 1-4). Sequencing pools were quantified on a Qubit 2.0 fluorometer (Invitrogen Inc., Waltham, MA) to ensure a 4nM starting concentration for sequencing preparations on a NovaSeq 6000 and NextSeq 550 sequencing platforms. Sequence pools destined for the NovaSeq 6000 loaded onto an S4 flowcell (each lane loaded with 1 sequencing pool of 384 samples) following the manufacturer’s protocol for the Xp-loading workflow (Illumina Inc., San Diego, CA.). For sequencing performed on the NextSeq 550 platforms, 1 sequencing pool (384 samples) loaded onto a High Output flowcell following the manufacturer’s protocol (Illumina Inc., San Diego, CA.). Across all sequencing platforms parameters were as follows: 2 x 75bp paired-end, dual-indexed reads. Approximately 5,000 samples were run on the NovaSeq 6000 and 15,000 on the NextSeq 550.

Data Analysis

FASTQ files were generated for each sample after demultiplexing of the raw sequencing data on Base Space Sequence Hub (BSSH). Detection of the virus, generation of the consensus sequence and lineage/clade determination was performed using the DRAGEN COVID Lineage App v3.5.1 and v3.5.2 on BSSH.

Statistical Analysis

Spearman's rank correlation coefficient, a rank-based analogue of Pearson's correlation, was used to measure the possibly non-linear but still monotonic relationship between mean Ct and % Genome Coverage. Tukey's test was used to test for differences in mean Ct value among the three collection methods.

Viral Detection

Each sample was evaluated for the presence of the virus in the sequencing data using a k-mer based algorithm prior to performing variant calling steps. Each read was broken down into all possible contiguous 32 bases segments (32-mers) and compared to a pre generated list of 32-mers from all the amplicons in the CovidSeq assay (98 from SARS-CoV-2 and 11 from human control genes). An amplicon was considered detected when at least 150 k-mers were matched between the sample and reference. When 5 or more SARS-CoV-2 amplicons were detected, the virus was recorded as present in the sample and the variant calling pipeline was invoked.

Alignment, Variant Calling and Consensus

Reads were aligned against the NC_045512.2 sequence (Severe acute respiratory syndrome coronavirus 2 isolate Wuhan-Hu-1) using the DRAGEN Map/Align algorithm. Reads pairs that aligned to the same start and end positions were marked as duplicate by default (only the highest quality pair is retained for variant calling). No quality pre-filtering of the data was performed by default, but portions of reads that do not match the reference were soft clipped by the aligner (removed for variant calling but retained in the BAM file) to eliminate mismatches due to a drop in read quality or chimeric artifacts from the PCR. For variant calling, the DRAGEN Somatic default down sampling parameters were used (no more than 10,000 reads per 300 base window around a variable position and no more than 50 reads starting at the same position were considered, random removal with fixed seed for reproducibility). Each variant was then assigned a somatic quality (SQ) score and marked as ‘weak_evidence’ if falling below a fixed threshold (SQ<3.0 hard filtering). Variant calling results were saved in a VCF formatted file and a consensus sequence generated using the bcftools CONSENSUS command. Any base covered with less than 10 reads was assigned an N (hard-masking) and any variable base was assigned the major allele base by default. Leading and trailing masked bases were then removed from the consensus by default.

Lineage and Clade Assignment

Consensus sequences were analyzed with the Pangolin and NextClade pipelines to generate the lineage and clade information, respectively. By default, the latest version of the PANGOLearn and NextClade databases were downloaded for the most up to date lineage and clade information.

Data Delivery to CDC

Data delivery to CDC was performed through our internal web application. Along with connecting to the BaseSpace file system, the application also connected to the CDC S3 data bucket using standard AWS S3 command line connection protocol using key based authentication1. Data transfer was performed on a per sample basis per CDC guidelines. For each sample the application created a file stack with pertinent files for samples that pass the quality thresholds. The application then opened a connection to the CDC S3 bucket and created a dated folder. Batch folders were created inside the dated primary folder when more than one batch of transfers were processed on a given day. One subfolder per sample was then created in the batch folder. The application then transferred all the pertinent files for the sample into the sample folder and continued this process until all sample data were transferred. The application created a cumulative metadata file containing the metadata associated with the samples transferred and transferred the file to the S3 bucket.

Sample Characteristics

A total of 24,441 SARS-CoV-2 positive samples (April to June of 2021) were sequenced (Table S1). Of these, the majority (24,237) samples were saliva collections (SA), 131 and 73 were from nasopharyngeal swabs (NP) and oropharyngeal swab collections (OC), respectively (Table 1). Of the 24, 441 individuals, 74% were from NJ and MN, with the remaining 26% distributed across the remaining 50 states. Where data was provided (21,125 patient samples) the mean age was 33 (ranging from 0-97) with a gender distribution (20,868 patients) of 38% females, 46% males and 15% with no gender data available (Table 1, Table S1). Of the 24,441 positive individuals, ~60% self-identified with an ethnicity, ~10% Hispanic or Latino, ~0.02%, Black not of Hispanic Origin, ~50% Non-Hispanic or Latino and White not of Hispanic Origin ~0.13% and 14% were either other or unknown origin. The remaining declined to identify, or information was unavailable. Vaccination status was provided by 23,329 (~95%) of patients. Of those ~7% indicated yes while 88% indicated no, the remaining declined to provide status.

Table 1

Totals and Percentages of Samples Sequenced and Characteristics
Total Samples	24,441
Sample Type
Saliva	24,237
Nasopharyngeal	131
Oropharyngeal	73
Geographic Location
MN	16,485
NJ	1,684
CA	386
Other U.S. States and Territories	5,886
Avg Age Yrs (Range)
Males	32.35 (0-94)
Females	32.17 (0-99)
Undisclosed	34.23 (6-74)
Mean	32.29 (0-99)
Gender (% Samples Sequenced)
Male	11,418 (46.72)
Female	9,450 (38.66)
Not Identified	3,573 (14.62)
Ethnicity (% Samples Sequenced)
Hispanic or Latino	2,518 (10.30)
Non-Hispanic or Latino	12,112 (49.56)
Others/Unknown Not Disclosed	9,811 (40.14)
Vaccination Status (% Yes/No/Unknown)
Yes	1,757 (7.18)
No	21,572 (88.26)
Unknown	1,113 (4.55)

Genome Coverage and Ambiguity Rates Between Sequencing Instruments

Of all the samples, 73.5% (17,974) samples were sequenced on the NextSeq 550, while 26.46% (6,467) were sequenced on the NovaSeq 6000. The average sequencing coverage between the two was ascertained using average sequencing coverage over the SARS-COV2 genome and fraction of nucleotides masked due to sequencing ambiguity in the consensus sequence generated. The global average SARS-CoV2 genome coverage for all samples sequenced across both instruments was 1324x. Samples run on the NovaSeq 6000 had twice the average coverage (2133x) compared to samples run on NextSeq550 (1034x) because more reads per sample were generated on the NovaSeq 6000. The fraction of masked nucleotides in consensus sequence generated was globally 5.9%. The ambiguity fraction rate using the NovaSeq6000 was 2.9% while genomes sequenced on the NextSeq550 had a higher ambiguity rate at 6.9% (Figures 2a, b and c).

Sequence Data Quality Between Sample Collection Methods

The mean RT-PCR Ct values for the three sample types used in this study across NP, OP and SA were 20.83, 21.83 and 22.70 respectively. By Tukey's test, the mean Ct values differed significantly between SA and NPC (p = 4e-7), but not between the other sample pairs. With respect to SARS-CoV2 sequence data quality metrics, several sample specific patterns were identified. The rate of failure of analytical detection of SARS-CoV2 sequence was highest in OP samples (4.11%, 3 of 73 samples), followed by SA samples (1.58%, 382 of 24, 237 samples), The sample type exhibiting the lowest rate of SARS-CoV2 detection failure was NP (0%, 0 of 131 samples). An important measure of input sample and sequence data quality is reflected in the percentage of consensus sequence masked as ambiguous. Higher values of this percentage are indicative of greater proportion of data that is not informative in variant determination. Figure 3a depicts the average percentage of consensus sequence masked as ambiguous for the three sample types used in this study. With a value of 0.5%, NPC samples yielded the highest quality sequence data, while SA (5.84%) and OPC swabs (41%) performing relatively less well. Another critical measure of quality, sequencing depth, was assessed for the three sample types as depicted in Figure 3b. This analysis determined that NPC and SA samples yielded the greatest depth, exhibiting values of 1,948x and 1,323x, respectively, while OPC sample coverage was the lowest at 586x.

Association Between Sequencing-Based Detection of SARS-CoV2 Virus and Baseline RT- PCR Ct Values

Samples were selected for inclusion in the sequencing study based on a minimum threshold of SARS-CoV2 detection by RT-PCR (N Gene Ct value threshold ≤ 30). Further investigation of the SARS-CoV2 PCR data of the samples in this study reveals a direct correlation between the ability to detect SARS-COV2 viral content by sequencing and the Ct values of the N gene. This observation held across all three sample types. Of the 24,441 samples sequenced, 385 (1.5%) were negative for SARS-CoV2 virus by sequencing, while 24,056 samples were positive. Among the negative samples, the mean Ct value for the N Gene was 24.5, while for positive samples it was 22.5 (Figure 4a, Table 2).

Table 2

Mean Ct Value Ranges for Three SARS-CoV-2 Genes in Detected and Not Detected Samples
	# Samples	x̅ of N Gene Ct	x̅ of S Gene Ct	x̅ of ORF1Ab Ct
Not Detected	385	24.56	24.75	24.79
Detected	24,056	22.57	22.47	22.64
Total	24,441	22.60	22.54	22.68

Association between Ct Value and Genome Coverage

We observed a strong inverse association between higher coverage and N gene Ct values as determined by RT-PCR (partial r, controlling for sequencer, = -0.58, partial r- squared = 0.34, p < 1e-15) (Figure 5) with higher Ct values resulting in lower overall mean genomic coverage. Spearman's rank correlation coefficient between Mean Ct and % Genome Coverage was -0.52 (p < 1e-15) for samples sequenced on the NovaSeq and -0.75 (p < 1e-15) for samples sequenced on the NextSeq. To examine the relationship between baseline Ct value and mean genome coverage, we divided the samples into 4 groups, group 1 with Ct values <22, group 2 with Ct values between 22 and 25, group 3 with Ct values between 25 and 28, and group 4 with Ct values >28. We saw a clear difference in mean genome coverage among the four groups (Figure 5). Group 1, with lowest Ct samples had the highest genomic coverage while group 4 with highest Ct values, had the lowest mean genomic coverage. We identified only 8 samples in our dataset that had mean genome coverage less than 100x. Of the remaining 24,433 samples we determined that a mean Ct value of 26.5 for NextSeq 550 and 27.9 for NovaSeq 6000 was a threshold for producing high quality genome sequencing reads (Figure 5a, b and c).

Strain Prevalence

We identified a total of 161 lineages of SARS-CoV2 variants in our dataset. The 10 most prevalent lineages are listed in Table 3. For a full list of all variants identified refer to Supplementary Table S1. The most prevalent lineage identified was B.1.1.7 or the Alpha variant (65%, n=15, 806), followed by B.1.526 variant (5.54%, n=1,330). We identified 3 out of the 4 CDC variants of interest (B.1.525, B.1.526, B.1.617.1) and all CDC variants of concern B.1.1.7, B.1.351, B.1.351.2, B.1.351.3, B.1.617.2, P.1 and P.1 sublineages) in our study (Figure 6a). The only CDC variant of interest that was not identified in our study within the sample set and time frame tested was B.1.617.3. We observed a rapid increase in the B.1.617 and AY lineages during May and June 2021 in our dataset, while the Alpha variant by proportion declined during that same period of time. Further details on trends in lineage discovery in our study as the data continued to be accumulated are provided in supplementary table 1.

Table 3

**Incidence of CDC SARS-Cov2 Variants of Concern in Population of Samples Sequenced**
Lineage	#Samples	% Of total Sequenced
B.1.1.7	15,806	65.86
B.1.526	1,330	5.54
B.1.429	1,015	4.23
B.1.2	950	3.96
B.1.1.519	730	3.04
B.1.526.2	488	2.03
P.1	435	1.81
B.1.427	411	1.71
B.1	396	1.65

Genomics-based SARS-CoV-2 surveillance is an important tool for monitoring viral transmission during the current and future phases of the COVID-19 pandemic. In this work, we sequenced the genomes from >24,000 SARS-CoV-2-positive samples collected during routine diagnostic testing. Furthermore, we analyzed the genomic epidemiology and sequencing applications of SARS-CoV-2 for the CDC surveillance program between March 2021 and June 2021, with a focus on variant and lineage determination, quality of sequencing results between sequencing instruments and the association between Ct values, genome coverage and variant detection.

Various NGS-based approaches have been developed to perform SARS-CoV-2 WGS using different sequencing platforms 12,13. These include direct RNA sequencing and metagenomics, amplicon-based methods and oligonucleotide capture based methods. We employed COVIDseq for mass-scale SARS-CoV-2 genomic surveillance 14 and demonstrated that COVIDseq enables near-complete coverage of the SARS-CoV-2 genome. We used statistical analysis to demonstrate that there were differences in average coverage between the sequencing instruments with the NovaSeq 6000 having larger output per sample compared to the NextSeq 550. In addition, we determined that a Ct value of 26.5 for NextSeq and 27.9 for NovaSeq served as thresholds for producing high quality genomes; above these thresholds sequencing quality degraded. High quality genomes were identified as ones that have a fraction of ambiguous nucleotides in the consensus sequence of less than 10% and average genome coverage > 100x. Sequencing of samples with Ct values > 35 has previously been reported to show a sizable fraction of the reads are aligned to the human genome, independently of the method used to prepare the libraries, resulting in inconsistency in lineage and variant detections

14.

We demonstrate that reports regarding variants of concern, such as transmission of the Delta variant, are consistent with other US and international reports 15. We identified a total of 161 lineages of the SARS-CoV2 variants in our dataset and the top 10 lineages were consistent with data reported to date beyond the present study. Moreover, with the increased number of positive cases in summer and winter of 2021, these data may also be consistent with vaccine breakthrough infections 15. However, additional studies are warranted. Additional variants, such as the Delta variant, where the variant of interest may not Although the Delta variant continues to represent the majority of SARS-CoV2 infections, sequencing data between the months of June and September 2021 are demonstrated increasing rates of transmission of another variant of concern, the Mu (B.2.621) variant, first identified in Colombia in January of 2021 16. Although the Mu variant accounts for only about 0.1% of cases within the present dataset, 77 positive samples with B.1.621 (Mu) have been identified since June of 2021. For data generated between July and October 2021, see Supplemental Table 1.

By comparing the RT-PCR Ct values with the ability to detect variants of SARS-CoV2, a likely correlation with the clinical features of the infection may be inferred. Increased genome coverage associated with lower Ct values are likely due to higher viral load in each sample 17. These results are consistent with previous reports where the ability to accurately detect variants correlates with lower Ct values 18. However, most of these samples are from SA-based collections. Conflicting evidence regarding the sensitivity between various collection methods for detecting SARS-CoV2 positive patients have been previously reproted 6,19,20. Thus comparison between SA and other collections method remains to be performed. The data presented in this study indicate that NP and SA are the optimal sample types for SARS-CoV2 surveillance sequencing studies, both in terms of strain identification and sequencing depth of coverage. Additional data and studies will need to be performed to further elucidate sensitivities associated with collection methods. As demonstrated by the data above (Table 1), most samples tested derived from patients who indicated they were not vaccinated at the time of testing. Of those who were vaccinated the prominent variant in those breakthrough infections (April-June 2021) was the B.1.1.7 (UK) variant. During this same period 28 cases of the Delta (B.1.617) variant were identified. However, as demonstrated in Figure 6, from June-July 2021 the occurrence of the UK variant dramatically decreases with a significant rise in the Delta variant that also correlated with an increase is positive cases throughout the U.S. and is now the prominent variant identified throughout the U.S. Vaccination status of samples sequenced from July through the remainder of 2021 are currently being collected (data not published). Thus, additional studies and analyses will be required to correlate, vaccination status, viral loads, sequencing variant impacts and clinical severity. The incomplete demographic data of all samples in aggregate which were sequenced limits the ability to stratify variant transmission between not only geographical regions but also within gender, age and other factors that may contribute to increased transmission rate. A recent and potentially more transmissible variant was detected first in South Africa B.1.1.529 (Omicron) 21. This variant has also been detected in Canada and Europe. However, at the time of submission of this manuscript this variant has not been reported within the U.S. through sequencing efforts. It is anticipated that detection within the U.S. will occur in the near term. Of interest is the concern that patients positive for this variant may be missed as the variant results in an S gene drop out. For these reasons it is imperative the appropriate methods for detection of positive SARS-CoV2 patients can detect the N and Orf genes as well.

In summary, the CDC surveillance screening program for SARS-CoV-2 variant transmission using whole viral genome sequencing is an important approach for population- based surveillance and control of viral transmission in the next phase of the COVID-19 pandemic. As SARS-CoV-2 strains continue to be sequenced by government, private, and academic entities all over the world, the sequencing data must continue to be shared publicly. These sequencing efforts will allow genomic-based surveillance of the virus and data sharing and contribute to efforts to understand population-level spread and control of the SARS-CoV-2 virus.

Ethics approval and consent to participate:

This study was determined to be exempt under 45 CFR § 46.104(d)(4), and exempt from informed consent because the research involves the use of identifiable private information/biospecimens; and information, which may include information about biospecimens, is recorded by the investigator in such a manner that the identity of the human subjects cannot readily by ascertained directly or through identifiers linked to the subjects, the investigator does not contact the subjects, and the investigator will not re-identify subjects (WCG IRB Work Order #1-1493985-1).

Consent for publication:

Not applicable

Availability of data and materials:

The consensus sequences were uploaded to GISAID (https://www.gisaid.org/) The accession numbers for each sample are available in GISAID database

Competing interests: Not applicable

Funding:

Centers for Disease Control and Prevention Contract No. 75D30121C10587

Authors’ contributions:

C.G.: conceptualization, formal analysis, investigation, visualization, writing—original draft; M.S.: conceptualization, formal analysis, investigation, visualization, writing—original draft; C.B.: conceptualization, formal analysis, investigation, visualization, writing—original draft; M.K: methodology, investigation, writing—original draft; Y.W: methodology, investigation,—original draft; J.S.: methodology, investigation-original draft; J.M: methodology; J.L.: supervision, methodology; E.K.: supervision, methodology; S.B.: Statistical Analysis, conceptualization, writing—original draft; D.G: supervision; G.B, Supervision, funding acquisition, review and editing; A.W.P. Sample acquisition, writing—review and editing; M.P.: funding acquisition, writing—review and editing; A.B.: funding acquisition, writing—review and editing; F.C.R.: funding acquisition, writing—review and editing; R.H.: funding acquisition, writing—review and editing; R.G.: funding acquisition, writing—review and editing; S.A.N; conceptualization, formal analysis, investigation, visualization, writing—original draft , funding acquisition, writing—review and editing. All authors read and approved the final manuscript.

Acknowledgements:

The authors wish to thank the Centers for Disease Control for guidance and wide base of knowledge making this study possible. for the team efforts and the wide base of knowledge making this study possible. The authors wish to thank the New Jersey and Minnesota IBX qPCR and Sequencing teams for contributing to the testing and sequencing efforts throughout these studies.

https://www.cdc.gov/mmwr/volumes/70/wr/mm7003e2.htm.
Hoffmann M, Arora P, Gross R, et al. SARS-CoV-2 variants B.1.351 and P.1 escape from neutralizing antibodies. Cell. 2021;184(9):2384-2393e2312.
Liu J, Liu Y, Xia H, et al. BNT162b2-elicited neutralization of B.1.617 and other SARS- CoV-2 variants. Nature.2021;596(7871):273-275.
Planas D, Bruel T, Grzelak L, et al. Sensitivity of infectious SARS-CoV-2 B.1.1.7and B.1.351 variants to neutralizing antibodies. Nat Med. 2021;27(5):917-924.
Zhou D, Dejnirattisai W, Supasa P, et al. Evidence of escape of SARS-CoV-2variantB.1.351 fromnatural and vaccine-induced sera. Cell. 2021;184(9):2348-2361e2346.
Wise J. Covid-19: New coronavirusvariant is identified in UK. BMJ. 2020;371:m4857.
Bian L, Gao Q, Gao F, et al. Impact of the Delta variant on vaccine efficacy and response strategies. Expert Rev Vaccines.2021:1-9.
https://www.cdc.gov/coronavirus/2019-ncov/variants/cdc-role-surveillance.html.
https://www.fda.gov/media/137776/download.
Bhoyar RC, Jain A, Sehgal P, et al. High throughput detection and geneticepidemiology of SARS-CoV-2 using COVIDSeq next-generation sequencing. PLoS One. 2021;16(2):e0247115.
Bhoyar RC, Senthivel V, Jolly B, et al. An optimized, amplicon-based approach for sequencing of SARS-CoV-2 from patient samples using COVIDSeq assay on Illumina MiSeq sequencing platforms. STAR Protoc. 2021;2(3):100755.
Nasir JA, Kozak RA, Aftanas P, et al. A Comparison of Whole Genome Sequencing of SARS-CoV-2 Using Amplicon-Based Sequencing, Random Hexamers, and Bait Capture. Viruses. 2020;12(8).
Pillay S, Giandhari J, Tegally H, et al. Whole Genome Sequencing of SARS-CoV-2: Adapting Illumina Protocols for Quick and Accurate Outbreak Investigation during a Pandemic. Genes (Basel).2020;11(8).
Simonetti M, Zhang N, Harbers L, et al. COVseq is a cost-effective workflow for mass- scale SARS-CoV-2 genomic surveillance. Nat Commun.2021;12(1):3903.
Farinholt T, Doddapaneni H, Qin X, et al. Transmission event of SARS-CoV-2 delta variant reveals multiple vaccine breakthrough infections. BMC Med.2021;19(1):255.
Laiton-Donato K, Franco-Munoz C, Alvarez-Diaz DA, et al. Characterization of the emerging B.1.621 variant of interest of SARS-CoV-2. Infect Genet Evol. 2021;95:105038.
Rabaan AA, Tirupathi R, Sule AA, et al. Viral Dynamics and Real-Time RT-PCR Ct Values Correlation with Disease Severity in COVID-19. Diagnostics (Basel).2021;11(6).
Roquebert B, Haim-Boukobza S, Trombert-Paolantoni S, et al. SARS-CoV-2 variants of concern are associated with lower RT-PCR amplification cycles between January and March 2021 in France. Int J Infect Dis.2021;113:12-14.
Congrave-Wilson Z, Lee Y, Jumarang J, et al. Change in Saliva RT-PCR Sensitivity Over the Course of SARS-CoV-2 Infection. JAMA.2021;326(11):1065-1067.
Tan SH, Allicock O, Armstrong-Hough M, Wyllie AL. Saliva as a gold-standard sample for SARS-CoV-2 detection. Lancet Respir Med.2021;9(6):562-564.
https://www.who.int/news/item/26-11-2021-classification-of-omicron-(b.1.1.529)-sars-cov-2-variant-of-concern

GoswamietalSupplementaryTable1.xlsx

Download PDF

Reviews received at journal
26 Feb, 2022
Reviewers invited by journal
23 Feb, 2022
First submitted to journal
04 Dec, 2021
Editorial decision: Minor revision
03 Dec, 2021

You are reading this latest preprint version

Identification of SARS-CoV-2 Variants Using Viral Sequencing for the Centers for Disease Control and Prevention Genomic U.S. Surveillance Program.

Status:

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

Background

Methods

Program Selection

Patients and sample acquisition

Criteria for SARS-CoV-2 Viral Sequencing

RNA isolation

Real-Time PCR for SARS-CoV-2

Library Preparation and Sequencing

Data Analysis

Statistical Analysis

Viral Detection

Alignment, Variant Calling and Consensus

Lineage and Clade Assignment

Data Delivery to CDC

Results

Discussion

Conclusion

Declarations

References

Supplementary Files

Status:

Version 1