Rapidly Identifying New Coronavirus Mutations of Potential Concern in the Omicron Variant Using an Unsupervised Learning Strategy (preprint)

Extensive mutations in the Omicron spike protein appear to accelerate the transmission of SARS-CoV-2, and rapid infections increase the odds that additional mutants will emerge. To build an investigative framework, we have applied an unsupervised machine learning approach to 4296 Omicron viral genomes collected and deposited to GISAID as of December 14, 2021, and have identified a core haplotype of 28 polymutants (A67V, T95I, G339D, R346K, S371L, S373P, S375F, K417N, N440K, G446S, S477N, T478K, E484A, Q493R, G496S, Q498R, N501Y, Y505H, T547K, D614G, H655Y, N679K, P681H, N764K, K796Y, N856K, Q954H, N69K, L981F) in the spike protein and a separate core haplotype of 17 polymutants in non-spike genes: (K38, A1892) in nsp3, T492 in nsp4, (P132, V247, T280, S284) in 3C-like proteinase, I189 in nsp6, P323 in RNA-dependent RNA polymerase, I42 in Exonuclease, T9 in envelope protein, (D3, Q19, A63) in membrane glycoprotein, and (P13, R203, G204) in nucleocapsid phosphoprotein. Using these core haplotypes as reference, we have identified four newly emerging polymutants (R346, A701, I1081, N1192) in the spike protein (p-value=9.37*10 -4 , 1.0*10 -15 , 4.76*10 -7 and 1.56*10 -4 , respectively), and five additional polymutants in non-spike genes (D343G in nucleocapsid phosphoprotein, V1069I in nsp3, V94A in nsp4, F694Y in the RNA-dependent RNA polymerase and L106L/F of ORF3a) that exhibit significant increasing trajectories (all p-values < 1.0*10 -15 ). In the absence of relevant clinical data for these newly emerging mutations, it is important to monitor them closely. Two emerging mutations may be of particular concern: the N1192S mutation in spike protein locates in an extremely highly conserved region of all human coronaviruses that is integral to the viral fusion process, and the F694Y mutation in the RNA polymerase may induce conformational changes that could impact Remdesivir binding.

Variants in nucleocapsid protein and endoRNase are found to associate with severe COVID-19 in a case-control study in Washington State, USA (preprint)

SARS-CoV-2 is spreading worldwide with continuously evolving variants, some of which occur in the Spike protein and appear to increase the viral transmissibility. However, variants that cause severe COVID-19 or lead to other breakthroughs have not been well characterized. To discover such viral variants, we assembled a cohort of 683 COVID-19 patients; 388 inpatients (“cases”) and 295 outpatients (“controls”) from April to August 2020 using electronically captured COVID test request forms and sequenced their viral genomes. To improve the analytic power, we accessed 7,137 viral sequences in Washington State to filter out viral single nucleotide variants (SNVs) that did not have significant expansions over the collection period. Applying this filter led to the identification of 53 SNVs that were statistically significant, of which 13 SNVs each had 3 or more variant copies in the discovery cohort. Correlating these selected SNVs with case/control status, eight SNVs were found to significantly associate with inpatient status (q-values<0.01). Using temporal synchrony, we identified a four SNV-haplotype (t19839-g28881-g28882-g28883) which was significantly associated with case/control status (Fisher’s exact p=2.84*10 −11 ) that appeared in April 2020, peaked in June, and persisted into January 2021. This association was replicated (OR=5.46, p-value=4.71*10 −12 ) in an independent cohort of 964 COVID-19 patients (June 1, 2020 to March 31, 2021). The haplotype included a synonymous change N73N in endoRNase, and three non-synonymous changes coding residues R203K, R203S and G204R in the nucleocapsid protein. This discovery points to the potential functional role of the nucleocapsid protein in triggering “cytokine storms” and severe COVID-19 that led to hospitalization. The study further emphasizes a need for tracking and analyzing viral sequences in correlations with clinical status.

Towards Discovering SARS-CoV-2 Variants of High Consequence Based on Both Surveillance and Electronically Captured Health Data: First Year Experience in Washington State (January 2020-2021) (preprint)

Background: SARS-CoV-2 is continuously evolving with the emergence of variants of interest (VOI) or with variants of concern (VOC). While Variants of High Consequence (VOHC) are well defined, no such variants have been formally documented. Here we propose an integrated strategy and application towards discovering VOHC. Methods: We utilized 7,137 viral sequences collected from COVID-19 cases in Washington State from January 19, 2020 to January 31, 2021, to identify genome-wide viral single nucleotide variants (SNVs). Utilizing a non-parametric regression model, we selected a subset of SNVs that had significant and substantial expansions over the collection period. Further, using unsupervised learning, we identified multiple SNVs forming haplotypes. To evaluate their clinical relevance, we assembled a discovery cohort of COVID-19 cases (388 inpatients and 295 outpatients) to identify SNVs and haplotypes associated with hospitalization status, a proxy for disease severity. A logistic regression model was used to assess associations of SNVs with hospitalization status in the discovery cohort. These results were validated on an independent cohort of 964 genome sequences derived from COVID-19 cases in Washington State from June 1, 2020 to March 31, 2021. Finding: The analysis of the 7,137 sequences led to identification of 107 SNVs that were statistically significant (false positive error rate q-value <0.01) and substantial expansions (maximum value of locally averaged proportions, Pmax>0.10). Forty-one SNVs were considered urgent, because their SNV proportions persisted or expanded above 10% in January 2021, the last month of the current investigation period. Correlating with clinical data, eight SNVs were found to significantly associate with inpatient status (p-values<0.001). By their synchronized dynamics, two SNVs were haplotyped and the mutant haplotype (c15933t-g16968t) was observed among patients in the discovery cohort (Fisher’s exact p=1.53*10-10), and this association was validated in the validation cohort (OR=5.38, p=10-9). Similarly, a haplotype with 4 SNVs (t19839c-g28881a-g28882a-g28883c) was observed only among inpatients (p=1.53*10-10) in the discovery cohort. Discovered haplotypic association was validated in the independent validation cohort (OR=3.69, p-value=3.44*10-10) and was further validated after adjusting for sex, age and collection time (OR=5.46, p-value=4.71*10-12). Interpretation: The mutant haplotype t19839c-g28881a-g28882a-g28883c emerged in April 2020, remained undetected over eight months, and has now begun to re-emerge. Because of its strong association with hospitalization status and re-emergence, this mutant haplotype may be a candidate variant for VOHC, pending further investigation of a) its clinical association with the disease severity, b) asymptomatic transmissibility and/or c) immune evasion to approved vaccines. While preliminary, this result indicates the importance to conduct purpose-driven clinical follow up studies to discover and validate candidate variants for VOHC. Also of interest is the mutant haplotype c15933t-g16968t which expanded in May 2020 but subsided by October 2020. Due to its association with hospitalization, we recommend continued monitoring for re-emergence of this variant and further assessment of viral phenotype.Funding: National Institutes of Health grant R01-GM129325 National Institutes of Health/National Institute of Allergy and Infectious Diseases grant UM1 AI068635Declaration of Interest: The authors declare that they have no competing interests.Ethical Approval: This study was approved by the Human Subject Review Committee at Fred Hutchinson Cancer Research Center (IRB#6007-2043) and by the University of Washington Institutional Review Board (STUDY00000408).

Towards Discovering SARS-CoV-2 Variants of High Consequence Based on Both Surveillance and Electronically Captured Health Data: First Year Experience in Washington State (January 2020-2021) (preprint)

Background: SARS-CoV-2 is continuously evolving with the emergence of variants of interest (VOI) or with variants of concern (VOC). While Variants of High Consequence (VOHC) are well defined, no such variants have been formally documented. Here we propose an integrated strategy and application towards discovering VOHC.Methods: We utilized 7,137 viral sequences collected from COVID-19 cases in Washington State from January 19, 2020 to January 31, 2021, to identify genome-wide viral single nucleotide variants (SNVs). Utilizing a non-parametric regression model, we selected a subset of SNVs that had significant and substantial expansions over the collection period. Further, using unsupervised learning, we identified multiple SNVs forming haplotypes. To evaluate their clinical relevance, we assembled a discovery cohort of COVID-19 cases (388 inpatients and 295 outpatients) to identify SNVs and haplotypes associated with hospitalization status, a proxy for disease severity. A logistic regression model was used to assess associations of SNVs with hospitalization status in the discovery cohort. These results were validated on an independent cohort of 964 genome sequences derived from COVID-19 cases in Washington State from June 1, 2020 to March 31, 2021.Finding: The analysis of the 7,137 sequences led to identification of 107 SNVs that were statistically significant (false positive error rate q-value <0.01) and substantial expansions (maximum value of locally averaged proportions, Pmax>0.10). Forty-one SNVs were considered urgent, because their SNV proportions persisted or expanded above 10% in January 2021, the last month of the current investigation period. Correlating with clinical data, eight SNVs were found to significantly associate with inpatient status (p-values<0.001). By their synchronized dynamics, two SNVs were haplotyped and the mutant haplotype (c15933t-g16968t) was observed among patients in the discovery cohort (Fisher’s exact p=1.53*10-10), and this association was validated in the validation cohort (OR=5.38, p=10-9). Similarly, a haplotype with 4 SNVs (t19839c-g28881a-g28882a-g28883c) was observed only among inpatients (p=1.53*10-10) in the discovery cohort. Discovered haplotypic association was validated in the independent validation cohort (OR=3.69, p-value=3.44*10-10) and was further validated after adjusting for sex, age and collection time (OR=5.46, p-value=4.71*10-12). Interpretation: The mutant haplotype t19839c-g28881a-g28882a-g28883c emerged in April 2020, remained undetected over eight months, and has now begun to re-emerge. Because of its strong association with hospitalization status and re-emergence, this mutant haplotype may be a candidate variant for VOHC, pending further investigation of a) its clinical association with the disease severity, b) asymptomatic transmissibility and/or c) immune evasion to approved vaccines. While preliminary, this result indicates the importance to conduct purpose-driven clinical follow up studies to discover and validate candidate variants for VOHC. Also of interest is the mutant haplotype c15933t-g16968t which expanded in May 2020 but subsided by October 2020. Due to its association with hospitalization, we recommend continued monitoring for re-emergence of this variant and further assessment of viral phenotype.Funding Information: National Institutes of Health grant R01-GM129325. National Institutes of Health/National Institute of Allergy and Infectious Diseases grant UM1 AI068635Declaration of Interests: None to declare. Ethics Approval Statement: This study was approved by the Human Subject Review Committee at Fred Hutchinson Cancer Research Center (IRB#6007-2043) and by the University of Washington Institutional Review Board (STUDY00000408).

Tracking SARS-CoV-2 Spike Protein Mutations in the United States (2020/01 – 2021/03) Using a Statistical Learning Strategy (preprint)

The emergence and establishment of SARS-CoV-2 variants of interest (VOI) and variants of concern (VOC) highlight the importance of genomic surveillance. We propose a statistical learning strategy (SLS) for identifying and spatiotemporally tracking potentially relevant Spike protein mutations. We analyzed 167,893 Spike protein sequences from US COVID-19 cases (excluding 21,391 sequences from VOI/VOC strains) deposited at GISAID from January 19, 2020 to March 15, 2021. Alignment against the reference Spike protein sequence led to the identification of viral residue variants (VRVs), i.e., residues harboring a substitution compared to the reference strain. Next, generalized additive models were applied to model VRV temporal dynamics, to identify VRVs with significant and substantial dynamics (false discovery rate q-value <0.01; maximum VRV proportion > 10% on at least one day). Unsupervised learning was then applied to hierarchically organize VRVs by spatiotemporal patterns and identify VRV-haplotypes. Finally, homology modelling was performed to gain insight into potential impact of VRVs on Spike protein structure. We identified 90 VRVs, 71 of which have not previously been observed in a VOI/VOC, and 35 of which have emerged recently and are durably present. Our analysis identifies 17 VRVs ∼91 days earlier than their first corresponding VOI/VOC publication. Unsupervised learning revealed eight VRV-haplotypes of 4 VRVs or more, suggesting two emerging strains (B1.1.222 and B.1.234). Structural modeling supported potential functional impact of the D1118H and L452R mutations. The SLS approach equally monitors all Spike residues over time, independently of existing phylogenic classifications, and is complementary to existing genomic surveillance methods.