Preliminary in silico assessment of the specificity of published molecular assays and design of new assays using the available whole genome sequences of 2019-nCoV

biolaboro · January 31, 2020, 7:34am

Preliminary in silico assessment of the specificity of published molecular assays and design of new assays using the available whole genome sequences of 2019-nCoV

Authors:

Mitchell Holland¹, Daniel Negrón¹, Shane Mitchell¹, Mychal Ivancich¹, Katharine W. Jennings¹, Bruce Goodwin², and Shanmuga Sozhamannan^2.3

¹Noblis, Reston, VA 20191

²Defense Biological Product Assurance Office, Frederick, MD 21702

³Logistics Management Institute, Tysons, VA 22102

Acknowledgement: All the investigators/labs that published the sequence data at GISAID- also see table at the end

Similar analyses/References: Marion Koopmans; Initial assessment of the ability of published coronavirus primers sets to detect the Wuhan coronavirus: http://virological.org/t/initial-assessment-of-the-ability-of-published-coronavirus-primers-sets-to-detect-the-wuhan-coronavirus/321Source

WGS: 25 WGS sequences have been downloaded from GISAID; All other data from NCBI BLAST databases (nt, gss, and env_nt - last updated December 2019)

Attached zip file with data visualizations: nCoV_pset_report.zip (459.5 KB)

Analyses:

BioLaboro is an application for rapidly designing de novo assays and validating existing PCR detection assays. It is a user-friendly new assay discovery pipeline composed of three tools: BioVelocity®, Primer3, and PSET. BioVelocity® uses a rapid, accurate hashing algorithm to align sequencing reads to a large set of references (e.g. Genbank) (Sozhamannan et al., 2015). BioVelocity® creates a k-mer index to determine all possible matches between query sequences and references simultaneously using a large RAM system (i.e. an IBM Power8). This algorithm makes it possible to very quickly identify sequences conserved within or omitted from a set of target references. Primer3 (http://primer3.sourceforge.net/) is a tool for designing primers and probes for real-time PCR reactions. It considers a range of criteria such as oligonucleotide melting temperature, size, GC content, and primer-dimer possibilities. We use Primer3 along with our signature detection process to identify potential new primer sets. PSET (PCR Signature Erosion Tool) tests PCR assays in silico against the latest versions of public sequence repositories, or other reference datasets, to determine if primers and probes match only to their intended targets. Using this information, an assay provider can be better aware of potential false hits and be better prepared to design new primers when false hits become intractable.

Results:

The BioLaboro application detected four highly specific signature sequence regions that hit all 25 (available at the time of analysis) Wuhan genomes (Table 1). The detected signatures were found to occur in disparate locations on the genome (Figure 1). All four signatures were found to target all current Wuhan genomes, and three out of four of these signature regions did not sufficiently align to any known coronavirus or other organism in NCBI BLAST databases (Table 2).

Table 1. List of PCR assays evaluated in this analysis. First four assays newly created using BioLaboro to be specific to Wuhan coronavirus. Last four assays from Diagnostic detection of Wuhan coronavirus 2019 by real-time RTPCR by Corman et al 2020.

Identifier	length	forward	probe	reverse
2019-nCoV-noblis_1	165	TGATGGTGGTGTCACTCGTG	TGGTTTAGCCAGCGTGGTGGT	GAAGTGGGTTTTGTCGTGCC
2019-nCoV-noblis_2	168	GCCGCTGTTGATGCACTATG	ACGTGCTCGTGTAGAGTGTTTTGAT	ATGCATTGCCTGAGACGACA
2019-nCoV-noblis_3	272	CGGATGGCTTATTGTTGGCG	TGCTCGTTGCTGCTGGCCTT	TTGGCTTTGCTGGAAATGCC
2019-nCoV-noblis_4	218	TGTCGTTGACAGGACACGAG	TTCGTCCGTGTTGCAGCCGA	CGTACGTGGCTTTGGAGACT
ncov_e_gene	113	ACAGGTACGTTAATAGTTAATAGCGT	ACACTAGCCATCCTTACTGCGCTTCG	TGTGTGCGTACTGCTGCAATAT
ncov_n_gene	128	CACATTGGCACCCGCAATC	ACTTCCTCAAGGAACAACATTGCCA	CAAGCCTCTTCTCGTTCCTC
ncov_rdrp_1	100	GTGARATGGTCATGTGTGGCGG	CCAGGTGGWACRTCATCMGGTGATGC	TATGCTAATAGTGTSTTTAACATYTG
ncov_rdrp_2	100	GTGARATGGTCATGTGTGGCGG	CAGGTGGAACCTCATCAGGAGATGC	TATGCTAATAGTGTSTTTAACATYTG

Figure 1. Map of the Wuhan genome (NCBI Accession: MN908947.3) with assay signature locations (created using DNA Features Viewer Python library). Corman assays in blue, Noblis assays in red, and gene regions in green.

Table 2. Results from PSET analysis. The four new Noblis assays were compared alongside the four assays from Corman. Each assay was tested using Wuhan coronavirus (25 genomes) as the intended target. All off-target hits (PF, FP) are to entries in NCBI BLAST databases (nt, gss, and env_nt).

Assay			Confusion
identifier	type	target	PT	TP	TN	PF	FP	FN
2019-nCoV-noblis_1	Probe	Wuhan	22	3	415	NA	NA	NA
2019-nCoV-noblis_2	Probe	Wuhan	25	NA	691	NA	NA	NA
2019-nCoV-noblis_3	Probe	Wuhan	25	NA	656	NA	NA	NA
2019-nCoV-noblis_4	Probe	Wuhan	25	NA	356	NA	3	NA
ncov_e_gene	Probe	Wuhan	25	NA	75	353	15	NA
ncov_n_gene	Probe	Wuhan	25	NA	73	NA	339	NA
ncov_rdrp_1	Probe	Wuhan	NA*	25	169	433	85	NA
ncov_rdrp_2	Probe	Wuhan	NA*	25	586	1	61	NA
PT = Perfect True Positive = All assay components hit with 100% identity to the correct target
TP = True Positive = All assay components hit with >=90% identity over >=90% of the component length to the correct target
TN = True Negative = Partial hit to assay amplicon but insufficient assay component alignments to an incorrect target
PF = Perfect False Positive = All assay components hit with 100% identity to an incorrect target
FP = False Positive = All assay components hit with >=90% identity over >=90% of the component length to an incorrect target
FN = False Negative = Partial hit to assay amplicon but insufficient assay component alignments to the correct target
* The ncov_rdrp_1 and ncov_rdrp_2 assays have one mismatch in the reverse primer to all current Wuhan coronavirus genomes which prevents it from being called a perfect hit. This is an incorrect match between an ambiguous base code, “S”, and the reference sequence “T”, in the middle of the primer which will likely not affect binding. NA == 0.

Conclusions and Caveats:

This is preliminary analyses based on 25 sequences available at the time.

New sequences are added on an hourly basis and these Noblis signatures need to be tested against the new sequences to verify that no signature erosion is occurring, as described in Sozhamannan et al 2015 for Ebola sequences. These designs were generated entirely in silico , and have yet to be tested in the lab. Although the BioLaboro pipeline is designed on sound scientific principles and the results from analyses of Ebola and Lassa viruses using the in silico components have been demonstrated (Sozhamannan et al 2015 and Wiley et al 2019) these assays await validation before conclusions regarding their use for clinical testing can be made.

Our intent in publishing these nCoV real-time PCR assays is to make the community aware of the existence of these potential unique signature regions as well the availability of BioLaboro for rapid evaluation of existing assays and design of new assays.

References:

Diagnostic detection of Wuhan coronavirus 2019 by real-time RTPCR -Protocol and preliminary evaluation as of Jan 13, 2020- Victor Corman, Tobias Bleicker, Sebastian Brünink, Christian Drosten, Charité Virology, Berlin, Germany; Olfert Landt, Tib-Molbiol, Berlin, Germany; Marion Koopmans, Erasmus MC, Rotterdam, The Netherlands; Maria Zambon, Public Health England, London, Additional advice by Malik Peiris, University of Hong Kong; contact: christian.drosten@charite.de https://virologie-ccm.charite.de/en/
Sozhamannan, Shanmuga, et al. “Evaluation of signature erosion in Ebola virus due to genomic drift and its impact on the performance of diagnostic assays.” Viruses 7.6 (2015): 3130-3154.
Wiley, Michael R., et al. “Lassa virus circulating in Liberia: a retrospective genomic characterisation.” The Lancet Infectious Diseases 19.12 (2019): 1371-1378.

Appendix 1: Acknowledgement for 2019-nCoV genome sequences

The following table is from this blog at Virological.org: Phylogenetic analysis of 23 nCoV-2019 genomes, 2020-01-23; Phylogenetic analysis of 23 nCoV-2019 genomes, 2020-01-23.

Table 3 . nCoV2019 genome sequences used in this analysis, the GISAID 6 accession numbers and submitting labs.

GISAID Accession	Strain	Location	Collection date	Lab
EPI_ISL_404227	BetaCoV/Zhejiang/WZ-01/2020	Zhejiang, China	2020-01-16	1
EPI_ISL_404228	BetaCoV/Zhejiang/WZ-02/2020	Zhejiang, China	2020-01-17	1
EPI_ISL_402132	BetaCoV/Wuhan/HBCDC-HB-01/2019	China/Hubei Province	2019-12-30	2
EPI_ISL_402127	BetaCoV/Wuhan/WIV02/2019	China / Hubei Province / Wuhan City	2019-12-30	3
EPI_ISL_402128	BetaCoV/Wuhan/WIV05/2019	China / Hubei Province / Wuhan City	2019-12-30	3
EPI_ISL_402129	BetaCoV/Wuhan/WIV06/2019	China / Hubei Province / Wuhan City	2019-12-30	3
EPI_ISL_402130	BetaCoV/Wuhan/WIV07/2019	China / Hubei Province / Wuhan City	2019-12-30	3
EPI_ISL_403963	BetaCoV/Nonthaburi/74/2020	Thailand/ Nonthaburi Province	2020-01-13	4
EPI_ISL_403962	BetaCoV/Nonthaburi/61/2020	Thailand/ Nonthaburi Province	2020-01-08	4
EPI_ISL_402120	BetaCoV/Wuhan/IVDC-HB-04/2020	China / Hubei Province / Wuhan City	2020-01-01	5
EPI_ISL_402119	BetaCoV/Wuhan/IVDC-HB-01/2019	China / Hubei Province / Wuhan City	2019-12-30	5
EPI_ISL_402121	BetaCoV/Wuhan/IVDC-HB-05/2019	China / Hubei Province / Wuhan City	2019-12-30	5
EPI_ISL_402124	BetaCoV/Wuhan/WIV04/2019	China / Hubei Province / Wuhan City	2019-12-30	6
EPI_ISL_402123	BetaCoV/Wuhan/IPBCAMS-WH-01/2019	China / Hubei Province / Wuhan City	2019-12-24	7
EPI_ISL_402125	BetaCoV/Wuhan-Hu-1/2019	China	2019-12	8
EPI_ISL_403931	BetaCoV/Wuhan/IPBCAMS-WH-02/2019	China / Hubei Province / Wuhan City	2019-12-30	9
EPI_ISL_403928	BetaCoV/Wuhan/IPBCAMS-WH-05/2020	China / Hubei Province / Wuhan City	2020-01-01	9
EPI_ISL_403930	BetaCoV/Wuhan/IPBCAMS-WH-03/2019	China / Hubei Province / Wuhan City	2019-12-30	9
EPI_ISL_403929	BetaCoV/Wuhan/IPBCAMS-WH-04/2019	China / Hubei Province / Wuhan City	2019-12-30	9
EPI_ISL_403937	BetaCoV/Guangdong/20SF040/2020	Guangdong, China	2020-01-18	10
EPI_ISL_403936	BetaCoV/Guangdong/20SF028/2020	Guangdong, China	2020-01-17	10
EPI_ISL_403935	BetaCoV/Guangdong/20SF025/2020	Guangdong, China	2020-01-15	10
EPI_ISL_403934	BetaCoV/Guangdong/20SF014/2020	Guangdong, China	2020-01-15	10
EPI_ISL_403933	BetaCoV/Guangdong/20SF013/2020	Guangdong, China	2020-01-15	10
EPI_ISL_403932	BetaCoV/Guangdong/20SF012/2020	Guangdong, China	2020-01-14	10

[1] Department of Microbiology, Zhejiang Provincial Center for Disease Control and Prevention

[2] Hubei Provincial Center for Disease Control and Prevention

[3] Wuhan Institute of Virology, Chinese Academy of Sciences

[4] Department of Medical Sciences, Ministry of Public Health, Thailand & Thai Red Cross Emerging Infectious Diseases - Health Science Centre & Department of Disease Control, Ministry of Public Health, Thailand

[5] National Institute for Viral Disease Control and Prevention, China CDC

[6] Wuhan Institute of Virology, Chinese Academy of Sciences

[7] Institute of Pathogen Biology, Chinese Academy of Medical Sciences & Peking Union Medical College

[8] National Institute for Communicable Disease Control and Prevention (ICDC) Chinese Center for Disease Control and Prevention (China CDC)

[9] Institute of Pathogen Biology, Chinese Academy of Medical Sciences & Peking Union Medical College

[10] Department of Microbiology, Guangdong Provincial Center for Diseases Control and Prevention

kihohong · January 26, 2020, 10:05pm

I couldn’t match the primer from Lancet article (by Huang et al.) with the nCoV sequences.
Can anybody check this primer?

https://doi.org/10.1016/S0140-6736(20)30183-5
forward primer 5′-TCAGAATGCCAATCTCCCCAAC-3′;
reverse primer 5′-AAAGGTCCACCCGATACATTGA-3′;

biolaboro · January 26, 2020, 7:06pm

hhttps://www.thelancet.com/action/showPdf?pii=S0140-6736%2820%2930183-5

Reported in Methods section, in Procedures subsection on page 2 of article:

“The primers and probe target to envelope gene of CoV were used and the sequences were as follows: forward primer 5′-TCAGAATGCCAATCTCCCCAAC-3′; reverse primer 5′-AAAGGTCCACCCGATACATTGA-3′; and the probe 5′CY5-CTAGTTACACTAGCCATCCTTACTGC-3′BHQ1.”

The forward and reverse primers reported here do not seem to correspond to nCoV sequences. However, I found them in another paper listed below and are from Saffold Cardiovirus.

Saffold Cardiovirus in Children with Acute Gastroenteritis, Beijing, ChinaLili Ren, Richard Gonzalez, Yan Xiao, Xiwei Xu, Lan Chen, Guy Vernet, Gláucia Paranhos-Baccalà, Qi Jin, and Jianwei Wang

“Because VP1 genes of 2 SAFV-positive samples could not be amplified in this way, a newly designed primer pair (cardioVP1Fn: TCAGAATGCCAATCTCCCCAAC and cardioVP1Rn: AAAGGTCCACCCGATACATTGA) was used in combination with cardioVP1-2F/3R to amplify the VP1 gene based on the sequences obtained from our positive samples.”

Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 15, No. 9, September 2009

kihohong · January 27, 2020, 1:22am

I got it. Thank you so much for your nice explanation and nice work!
I checked Saffold virus sequences.
The primer pair perfectly matches to Saffold virus sequences.

biolaboro · January 28, 2020, 1:47am

PSET results updated with three new 2019-nCoV genomes; 28 total genomes. Also adding three assays for comparison from the CDC (https://www.cdc.gov/coronavirus/2019-ncov/downloads/rt-pcr-panel-primer-probes.pdf). New genomes have resulted in mismatches in previously perfect alignments for some assays.

Table 1. Results from PSET analysis. The four Noblis assays were compared alongside the four assays from Corman and three assays from the CDC. Each assay was tested using 2019-nCoV (28 genomes) as the intended target. All off-target hits (TN, PF, FP) are to entries in NCBI BLAST databases (nt, gss, and env_nt).

identifier	provider	PT	TP	TN	PF	FP	FN
2019-nCoV-noblis_1	Noblis	23	5	415	NA	NA	NA
2019-nCoV-noblis_2	Noblis	27	1	691	NA	NA	NA
2019-nCoV-noblis_3	Noblis	27	1	638	NA	NA	NA
2019-nCoV-noblis_4	Noblis	28	NA	356	NA	3	NA
ncov_e_gene	Corman et al	28	NA	75	353	15	NA
ncov_n_gene	Corman et al	27	NA	73	NA	339	1
ncov_rdrp_1	Corman et al	NA	28	169	433	85	NA
ncov_rdrp_2	Corman et al	1	27	586	1	61	NA
2019-nCoV_N1	CDC	27	1	381	NA	NA	NA
2019-nCoV_N2	CDC	27	1	371	NA	NA	NA
2019-nCoV_N3	CDC	28	NA	48	NA	346	NA

biolaboro · January 31, 2020, 8:42pm

PSET results updated with new 2019-nCoV genomes; 46 total genomes.

Table 1. Results from PSET analysis. The four Noblis assays were compared alongside the four assays from Corman and three assays from the CDC. Each assay was tested using 2019-nCoV (46 genomes) as the intended target. All off-target hits (TN, PF, FP) are to entries in NCBI BLAST databases (nt, gss, and env_nt).

identifier	provider	PT	TP	TN	PF	FP	FN
2019-nCoV-noblis_1	Noblis	36	10	415	NA	NA	NA
2019-nCoV-noblis_2	Noblis	45	1	689	NA	NA	NA
2019-nCoV-noblis_3	Noblis	45	1	637	NA	NA	NA
2019-nCoV-noblis_4	Noblis	46	NA	356	NA	3	NA
ncov_e_gene	Corman et al	46	NA	75	353	15	NA
ncov_n_gene	Corman et al	44	1	73	NA	339	1
ncov_rdrp_1	Corman et al	NA	46	169	433	85	NA
ncov_rdrp_2	Corman et al	1	45	586	1	61	NA
cdc_n1	CDC	44	2	381	NA	NA	NA
cdc_n2	CDC	45	1	371	NA	NA	NA
cdc_n3	CDC	45	1	48	NA	346	NA

Figure 1. Updated genome map showing locations of all assays. Map of the Wuhan genome (NCBI Accession: MN908947.3) with assay signature locations (created using DNA Features Viewer Python library). Corman assays in blue, Noblis assays in red, CDC assays in purple, and gene regions in green.

biolaboro · February 7, 2020, 7:45pm

PSET results updated with new 2019-nCoV genomes; 66 total genomes.

Table 1. Results from PSET analysis. The four Noblis assays were compared alongside the four assays from Corman and three assays from the CDC. Each assay was tested using 2019-nCoV (66 genomes) as the intended target. All off-target hits (TN, PF, FP) are to entries in NCBI BLAST databases (nt, gss, and env_nt).

identifier	provider	PT	TP	TN	PF	FP	FN
2019-nCoV-noblis_1	Noblis	47	19	417	NA	NA	NA
2019-nCoV-noblis_2	Noblis	65	1	688	NA	NA	NA
2019-nCoV-noblis_3	Noblis	65	1	682	NA	NA	NA
2019-nCoV-noblis_4	Noblis	65	NA	357	NA	3	1
ncov_e_gene	Corman et al	66	NA	76	353	15	NA
ncov_n_gene	Corman et al	64	1	74	NA	339	1
ncov_rdrp_1	Corman et al	NA	66	169	433	85	NA
ncov_rdrp_2	Corman et al	1	65	586	1	61	NA
cdc_n1	CDC	64	2	381	NA	NA	NA
cdc_n2	CDC	65	1	371	NA	NA	NA
cdc_n3	CDC	65	1	48	NA	346	NA

biolaboro · February 14, 2020, 8:55pm

PSET results updated with new 2019-nCoV genomes; 96 total genomes. Now just using the subset of genomes on GISAID marked as high quality. Sequence IDs tested in this analysis listed here: ncov_ids_tested.zip (386 Bytes)

Assays continue to perform well against new genome submissions. Only one assay, 2019-nCoV-noblis_4, showing potential false negatives. Assay 2019-nCoV-noblis_1 has just one mismatch in the probe for 27 genomes, resulting in less perfect matches, but still likely functional for all 96 genomes.

Table 1. Results from PSET analysis. The four Noblis assays were compared alongside the four assays from Corman and three assays from the CDC. Each assay was tested using 2019-nCoV (96 genomes) as the intended target. All off-target hits (TN, PF, FP) are to entries in NCBI BLAST databases (nt, gss, and env_nt).

identifier	provider	PT	TP	TN	PF	FP	FN
2019-nCoV-noblis_1	Noblis	69	27	372	NA	NA	NA
2019-nCoV-noblis_2	Noblis	96	NA	568	NA	NA	NA
2019-nCoV-noblis_3	Noblis	96	NA	439	NA	NA	NA
2019-nCoV-noblis_4	Noblis	94	NA	343	NA	3	2
ncov_e_gene	Corman et al	96	NA	42	353	15	NA
ncov_n_gene	Corman et al	95	1	55	NA	339	NA
ncov_rdrp_1	Corman et al	NA	96	75	433	87	NA
ncov_rdrp_2	Corman et al	NA	96	526	1	66	NA
cdc_n1	CDC	95	1	363	NA	NA	NA
cdc_n2	CDC	95	1	361	NA	NA	NA
cdc_n3	CDC	94	2	17	NA	346	NA

biolaboro · February 21, 2020, 8:51pm

PSET results updated with new 2019-nCoV genomes; 129 total genomes. Now just using the subset of genomes on GISAID marked as high quality. Sequence IDs tested in this analysis listed here: ncov_ids_tested.zip (437 Bytes)

Assays have more potential false negatives, although the majority of FNs (38 out of 42) are from samples collected from Pangolins. Only assays ncov_rdp_1 and cdc_n3 currently have no potential for FNs.

Table 1. Results from PSET analysis. The four Noblis assays were compared alongside the four assays from Corman and three assays from the CDC. Each assay was tested using 2019-nCoV (129 genomes) as the intended target. All off-target hits (TN, PF, FP) are to entries in NCBI BLAST databases (nt, gss, and env_nt).

identifier	provider	PT	TP	TN	PF	FP	FN
2019-nCoV-noblis_1	Noblis	91	32	372	NA	NA	6
2019-nCoV-noblis_2	Noblis	123	NA	568	NA	NA	6
2019-nCoV-noblis_3	Noblis	122	2	439	NA	NA	5
2019-nCoV-noblis_4	Noblis	121	5	343	NA	3	3
ncov_e_gene	Corman et al	127	1	42	353	15	1
ncov_n_gene	Corman et al	122	2	55	NA	339	5
ncov_rdrp_1	Corman et al	NA	129	75	433	87	NA
ncov_rdrp_2	Corman et al	NA	124	526	1	66	5
cdc_n1	CDC	122	2	363	NA	NA	5
cdc_n2	CDC	122	1	361	NA	NA	6
cdc_n3	CDC	121	8	17	NA	346	NA

biolaboro · February 28, 2020, 9:25pm

PSET results updated with new 2019-nCoV genomes; 152 total genomes. Now just using the subset of genomes on GISAID marked as high quality. Sequence IDs tested in this analysis listed here: ncov_ids_tested.zip (475 Bytes)

Table 1. Results from PSET analysis. The four Noblis assays were compared alongside the four assays from Corman and three assays from the CDC. Each assay was tested using 2019-nCoV (152 genomes) as the intended target. All off-target hits (TN, PF, FP) are to entries in NCBI BLAST databases (nt, gss, and env_nt).

identifier	provider	PT	TP	TN	PF	FP	FN
2019-nCoV-noblis_1	Noblis	106	39	373	NA	NA	7
2019-nCoV-noblis_2	Noblis	145	NA	568	NA	NA	7
2019-nCoV-noblis_3	Noblis	144	2	439	NA	NA	6
2019-nCoV-noblis_4	Noblis	144	5	343	NA	3	3
ncov_e_gene	Corman et al	149	2	42	353	15	1
ncov_n_gene	Corman et al	144	3	55	NA	339	5
ncov_rdrp_1	Corman et al	NA	152	75	433	87	NA
ncov_rdrp_2	Corman et al	NA	147	526	1	66	5
cdc_n1	CDC	144	3	363	NA	NA	5
cdc_n2	CDC	144	1	361	NA	NA	7
cdc_n3	CDC	143	9	17	NA	346	NA

biolaboro · March 6, 2020, 9:17pm

PSET results updated with new 2019-nCoV genomes; 190 total genomes. Now just using the subset of genomes on GISAID marked as high quality sampled from humans (pangolin samples removed). Sequence IDs tested in this analysis listed here: ncov_ids_tested.zip (534 Bytes)

Table 1. Results from PSET analysis. The four Noblis assays were compared alongside the four assays from Corman and three assays from the CDC. Each assay was tested using 2019-nCoV (190 genomes) as the intended target. All off-target hits (TN, PF, FP) are to entries in NCBI BLAST databases (nt, gss, and env_nt).

identifier	provider	PT	TP	TN	PF	FP	FN
2019-nCoV-noblis_1	Noblis	134	56	373	0	0	0
2019-nCoV-noblis_2	Noblis	188	2	568	0	0	0
2019-nCoV-noblis_3	Noblis	189	1	439	0	0	0
2019-nCoV-noblis_4	Noblis	184	0	343	0	3	6
ncov_e_gene	Corman et al	187	2	42	353	15	1
ncov_n_gene	Corman et al	189	1	55	0	339	0
ncov_rdrp_1	Corman et al	190	0	75	433	87	0
ncov_rdrp_2	Corman et al	190	0	526	1	66	0
cdc_n1	CDC	188	2	363	0	0	0
cdc_n2	CDC	189	1	361	0	0	0
cdc_n3	CDC	183	7	17	0	346	0

biolaboro · March 13, 2020, 9:58pm

PSET results updated with new 2019-nCoV genomes; 432 total genomes. Just using the subset of genomes on GISAID marked as high quality sampled from humans. Sequence IDs tested in this analysis listed here: 432_ids.zip (833 Bytes)

Noblis assay 4 showing some potential FNs, but most of these appear to be the result of gaps or sequencing errors for some sequences at the very start of the genome. This assay falls within the first 500bp of the genome. All other assays still performing very well in silico against new sequences.

Table 1. Results from PSET analysis. The four Noblis assays were compared alongside the four assays from Corman and three assays from the CDC. Each assay was tested using 2019-nCoV (432 genomes) as the intended target. All off-target hits (TN, PF, FP) are to entries in NCBI BLAST databases (nt, gss, and env_nt).

identifier	provider	PT	TP	TN	PF	FP	FN
2019-nCoV-noblis_1	Noblis	325	105	373	0	0	2
2019-nCoV-noblis_2	Noblis	422	10	568	0	0	0
2019-nCoV-noblis_3	Noblis	431	1	439	0	0	0
2019-nCoV-noblis_4	Noblis	409	4	343	0	3	19
ncov_e_gene	Corman et al	428	3	42	353	15	1
ncov_n_gene	Corman et al	431	1	55	0	339	0
ncov_rdrp_1	Corman et al	0	432	75	433	87	0
ncov_rdrp_2	Corman et al	0	432	526	1	66	0
cdc_n1	CDC	429	3	363	0	0	0
cdc_n2	CDC	430	2	361	0	0	0
cdc_n3	CDC	412	20	17	0	346	0

biolaboro · March 20, 2020, 10:10pm

PSET results updated with new 2019-nCoV genomes; 655 total genomes. Just using the subset of genomes on GISAID marked as high quality sampled from humans. Sequence IDs tested in this analysis listed here: 655_ids.zip (1.4 KB)

Previous Noblis assays showed some false negatives due to Ns, sequence gaps, and in one case the assay’s position at the very start of the genome. These have been replaced with five new assays generated at a later date using 96 complete genomes. The Noblis.57 assay and the German ncov_e_gene assay each have one FN that’s due to a stretch of Ns. All other assays still performing very well in silico against new sequences.

Table 1. Results from PSET analysis. The five Noblis assays were compared alongside the four assays from Corman and three assays from the CDC. Each assay was tested using 2019-nCoV (655 genomes) as the intended target. All off-target hits (TN, PF, FP) are to entries in NCBI BLAST databases (nt, gss, and env_nt).

identifier	provider	PT	TP	TN	PF	FP	FN
Noblis.12	Noblis	653	2	2	0	0	0
Noblis.40	Noblis	635	20	316	0	0	0
Noblis.42	Noblis	655	0	275	0	0	0
Noblis.44	Noblis	655	0	277	0	0	0
Noblis.57	Noblis	654	0	359	0	0	1
ncov_e_gene	Corman et al	651	3	42	353	15	1
ncov_n_gene	Corman et al	652	3	55	0	339	0
ncov_rdrp_1	Corman et al	0	655	75	433	87	0
ncov_rdrp_2	Corman et al	1	654	526	1	66	0
cdc_n1	CDC	647	8	363	0	0	0
cdc_n2	CDC	653	2	361	0	0	0
cdc_n3	CDC	628	27	17	0	346	0

Figure 1. Map of the SARS-CoV-2 genome (NCBI Accession: MN908947.3) with assay signature locations (created using DNA Features Viewer Python library). Noblis assays in red, Corman assays in blue, CDC assays in purple, and gene regions in green.

biolaboro · March 27, 2020, 8:20pm

PSET results updated with new 2019-nCoV genomes; 1620 total genomes. Just using the subset of genomes on GISAID marked as high quality sampled from humans. Sequence IDs tested in this analysis listed here: 1620_ids.zip (12.2 KB)

The Noblis.57 assay and the German ncov_e_gene assay still each have one FN that’s due to a stretch of Ns. All other assays still performing very well in silico against new sequences.

Table 1. Results from PSET analysis. The five Noblis assays were compared alongside the four assays from Corman and three assays from the CDC. Each assay was tested using 2019-nCoV (1620 genomes) as the intended target. All off-target hits (TN, PF, FP) are to entries in NCBI BLAST databases (nt, gss, and env_nt).

identifier	provider	PT	TP	TN	PF	FP	FN
Noblis.12	Noblis	1616	4	2	0	0	0
Noblis.40	Noblis	1546	74	316	0	0	0
Noblis.42	Noblis	1620	0	275	0	0	0
Noblis.44	Noblis	1619	1	277	0	0	0
Noblis.57	Noblis	1605	14	359	0	0	1
ncov_e_gene	Corman et al	1616	3	42	353	15	1
ncov_n_gene	Corman et al	1608	12	55	0	339	0
ncov_rdrp_1	Corman et al	0	1620	75	433	87	0
ncov_rdrp_2	Corman et al	1	1619	526	1	66	0
cdc_n1	CDC	1591	29	363	0	0	0
cdc_n2	CDC	1616	4	361	0	0	0
cdc_n3	CDC	1560	60	17	0	346	0