key: cord-0789953-nu0ei47n
authors: Salvatore, M.; Bhattacharyya, R.; Purkayastha, S.; Zimmermann, L.; Ray, D.; Hazra, A.; Kleinsasser, M.; Mellan, T. A.; Whittaker, C.; Flaxman, S.; Bhatt, S.; Mishra, S.; Mukherjee, B.
title: Resurgence of SARS-CoV-2 in India: Potential role of the B.1.617.2 (Delta) variant and delayed interventions
date: 2021-06-30
journal: nan
DOI: 10.1101/2021.06.23.21259405
sha: 6da4a08fa641e01f03f3dfb8077145eb90e0670a
doc_id: 789953
cord_uid: nu0ei47n

India has seen a surge of SARS-CoV-2 infections and deaths in early part of 2021, despite having controlled the epidemic during 2020. Building on a two-strain, semi-mechanistic model that synthesizes mortality and genomic data, we find evidence that altered epidemiological properties of B.1.617.2 (Delta) variant play an important role in this resurgence in India. Under all scenarios of immune evasion, we find an increased transmissibility advantage for B.1617.2 against all previously circulating strains. Using an extended SIR model accounting for reinfections and wanning immunity, we produce evidence in support of how early public interventions in March 2021 would have helped to control transmission in the country. We argue that enhanced genomic surveillance along with constant assessment of risk associated with increased transmission is critical for pandemic responsiveness.

Since the September peak, the incidence curve declined steadily to less than 10,000 daily new cases in February of 2021 (14) . National serosurveys and epidemiological model estimates indicated a substantial infection under-ascertainment rate, suggesting only about 6% of infections in India were reported by the end of 2020 (15) (16) (17) (18) . There were discussions about urban metros in India approaching herd immunity thresholds with many of them reporting more than 50% seropositivity (17, 19, 20) . The third national serosurvey in January 2021 indicated 21.5% of adults in India had evidence of a past COVID-19 infection (21) . As the country further relaxed restrictions, COVID-appropriate behaviors diminished with time (10) , crowded public transportation system restarted, large indoor and outdoor gatherings were taking place without meaningful adherence to proper face coverings. Social and religious events, weddings, political rallies, mass protests -all cultural facets that define the tapestry of life in India were in full force. The year 2021 started on an optimistic note with multiple vaccine trials going on in India and globally (10) with promising efficacy and safety results. India formalized operational guidelines for its national vaccine distribution with emergency-use approval for two vaccines (22) , including prioritizing beneficiaries (23) . Vaccination began on January 16 with an initial focus on healthcare workers. The vaccine roll-out in India has been sluggish, and initially this was in part due to an underappreciation of the threat that SARS-CoV-2 still posed. Only 8.7 million doses were administered in the first one month, and 0.65% of the population received at least one dose on February 15. By April 1, only 5.2% of the population had received at least one dose in India with less than 2% of the population fully vaccinated.

After a steady decline for about four months, an uptick in cases was noted in three Indian states in February 2021 (Maharashtra, Punjab and Chhattisgarh) with the national effective reproduction number crossing the threshold of one on February 14 (24) . No strict control measures were reintroduced in the two months following the initial indications of a resurgence in transmission. Indeed, the first comprehensive lockdown only started on April 14 in Maharashtra (25) , when India was already witnessing a staggering growth in infections. A massive humanitarian crisis unfolded that was termed a "national catastrophe", and calls were made for an international alliance and collaboration (26) .

Healthcare infrastructure collapsed under surges in hospitalizations. The extent of this collapse was such that various parts of India suffered from acute shortages of oxygen, steroidal treatment medicine (27) and testing kits. Crematoriums and burial grounds were overflowing (28, 29) . Multiple preprints and investigative reports suggest that the actual death toll far exceeds official numbers (30, 31) . In addition to lack of timely and stringent preventive measures guided by public health, emerging variants became a large part of the conversation around India's second wave (32, 33) .

In many other parts of the world, winter of 2020 brought resurgent transmission, with new SARS-CoV-2 variants being identified in the UK (Alpha/B.1.1.7), Brazil (Gamma/P.1) and

in South Africa (Beta/B.1.351) (34) , concurrent with a globally changing pandemic trajectory (35) . Many of these new variants were observed to have epidemiologically distinct changes in transmissibility as well as antigenic escape. In December 2020, the Ministry of Health and Family Welfare (MoHFW) in India launched a multi-laboratory genomic surveillance initiative formally referred to as the Indian SARS-CoV-2 Genome Sequencing Consortia (INSACOG) (36) to track the virus's evolution and identify new Variants of Concern (VOC). The year 2020 was a period of comparative evolutionary stasis for SARS-CoV-2 with infections attributed to the previously circulated lineages (37) . In this paper, we present an epidemiological analysis of the second wave of COVID-19 transmission in India. First, we compare the second wave to the first wave, nationally and across states and union territories, in terms of multiple public health metrics. Using a dynamic epidemiological model that integrates both genomic and COVID-19 mortality data, we then investigate the extent to which the emergence and altered epidemiological properties of the SARS-CoV-2 Delta variant (i.e., B.1.617.2 sub-lineage) might have driven the surge in the observed case and death counts in the second wave in India.

Finally, we estimate the number of deaths that could have been averted through an early nationwide intervention (like a lockdown) at various time points in March and April 2021 during the onset of the second wave. We conclude with some recommendations moving forward.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted June 30, 2021. ; https://doi.org/10.1101/2021.06.23.21259405 doi: medRxiv preprint

Comparison at a national level: We define the second wave as starting on February 14, 2021 , the day the effective reproduction number, Rt, exceeded one (i.e., Rt > 1), indicating a growth in case counts after 4 months of steady decline. For comparative purposes, we consider June 13, 2020, as the starting point of the first wave (Wave 1), the first time the daily case counts were comparable to the daily case counts observed on February 14, 2021, the start of the second wave (Wave 2). Our analysis period ends on May 31, 2021, for the descriptive analysis. As shown in Table 1 , Wave 2 is more severe than Wave 1 in nearly every metric of growth: higher maximum Rt, nearly 4 times higher maximum daily cases and deaths, and maximum daily test-positive rate (TPR) of 25% in Wave 2 versus 16% in Wave 1. There is growing consensus among epidemiologists that the second wave in India has had a greater severity than the first wave (44, 45) . While there have been more cases and deaths in Wave 2, there is a lower reported case-fatality rate compared to Wave 1 (1.0% vs 1.4%). This could partly be explained by reports that younger age groups were infected more in Wave 2 where the clinical infection fatality rate is lower, but this assertion is yet to be verified (46) . It is important to note that we do not have full follow-up data on Wave 2 and more deaths continue to be reported in June including some states like Bihar revising previously reported death numbers (47) . There exists a systemic delay in updating death records and underreporting of case and death counts appear to be a challenge with this data (31, 48) .

State-level comparisons: Multiple articles have now shown that there is substantial heterogeneity across Indian states and union territories, and the national data often masks this state-level variation (12). For this comparison, due to the differences in the . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) 3 .0. Only one union territory, namely the Andaman and Nicobar Islands, reports a lessthan-unity ratio of 0.9, indicating a less severe Wave 2 when compared to Wave 1. All other states and union territories return ratios that are greater than unity, suggesting a more severe Wave 2. Across all states and union territories, the median ratio of daily case rate for Wave 2 versus Wave 1 is 3.8. The states with the four highest case intensity ratios are (in decreasing order): Uttarakhand (5.6), Himachal Pradesh (5.3), Punjab (5.2) and

Gujarat (5.2) . These findings are reflected in Figure 1A .

Following a similar approach for death counts, Delhi reported 9,675 deaths in Wave 1 over 246 days, yielding a standardized death intensity of 39.3. In Wave 2, there were 13,348 deaths reported over 107 days in Delhi, yielding a standardized death rate of 124.8. Consequently, the ratio of daily death rates for Wave 2 versus Wave 1 in Delhi is 3.2. Two states report ratios that are less than unity (Tripura 0.7 and Odisha 0.99), indicating a less severe Wave 2 when compared to Wave 1, while all other states and union territories return ratios that are greater than unity. Across all other states and union territories, the median death rate ratio for Wave 2 versus Wave 1 is 3. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted June 30, 2021. ; https://doi.org/10.1101/2021.06.23.21259405 doi: medRxiv preprint Mizoram (7.9), Nagaland (7.2), Meghalaya (6.7) and Uttarakhand (6.6). These findings are reflected in Figure 1B . The exact numerical values are provided in Table S1 .

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. State level variation in standardised case (A) and death (B) intensity ratio of wave 2 to wave 1. Standardised case (death) intensity of a given wave for a specific region is defined as the cumulative count of cases (deaths) recorded in a given wave and region, divided by the duration (in days) of said wave. Wave 1 is defined as the period from June 13, 2020 -February 13, 2021 (246 days) and wave 2 is defined as the period from February 14, 2021 -May 31, 2021 (107 days).

Notes: 1. We define Wave 2 starting on February 14, 2021, the day the effective reproduction number, Rt exceeded unity after 4 months of steady decline, indicating a growth in case counts. For comparative purposes, we consider the part of Wave 1 that is comparable to Wave 2. Namely, we consider the start of Wave 1 on June 13, 2020, the first time the daily case counts were comparable to the daily case counts observed on February 14, 2021. 2. States with case (death) intensity ratios less than unity are shaded blue while those with higher ratios are shaded in increasingly darker shades of red.

3. Owing to insufficient space, ratios from some Union Territories have not been included in this figure.

The national peak occurred on September 16, 2020, with 97,860 daily cases during Wave 1 and on May 6, 2021, with 414,280 daily cases during Wave 2. However, we see Table S1a and S1b.

As shown in Figure 1 and Table S1 there is considerable spatial heterogeneity in infection burden and fatalities across states and union territories in India across the waves. The period just before and around the second wave has seen the introduction of multiple VOCs, and variants under investigation in India (49, 50) . This period is marked by a rapid increase in the infection share of B.1.617.2 which has, as of late April 2021 (51), become . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. in GISAID for Maharashtra is dated 10 th Dec 2020 (50) . The results presented in the main text are for an assumed IFR of 0.25% (31) and an under-reporting factor of 50% (suggesting 1 in 2 deaths are missed due to testing and other logistical constraints which is a very conservative estimate based on the literature (31)); results from a sensitivity analysis comprising eleven other scenarios are summarized in Table S4 . Table S2 for a list) -across the three plausible values of cross-infection (100% -no immunity escape; 75% -moderate possibility of reinfection; 50% -higher possibility of reinfection), our results consistently support the hypothesis of B.1.617.2 as being more transmissible than all previously . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. However, these results remain highly uncertain, particularly given the absence of reliable serological surveys to infer cumulative population infections to date and the IFR. Whilst we cannot quantify the exact extent and nature of the changes, our results highlight that immunity waning or relaxation of NPIs alone cannot explain the large second wave in Maharashtra. Instead, the results suggest that this was at least partially driven by the emergence of a variant with altered epidemiological properties. We would like to point out that our two-strain model doesn't account for NPIs explicitly but observed changes to the reproduction number as a consequence of the NPIs are modelled as a 7-day random walk (55) . We note that, our estimates of altered epidemiological properties for B. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 30, 2021. For lockdowns beginning in mid-April, the benefits are not as pronounced. Our findings indicate that a peak would have occurred shortly after a lockdown with a strong effect with around 150,000 cases at the peak. However, under a lockdown with moderate effect, a . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 30, 2021. ; https://doi.org/10.1101/2021.06.23.21259405 doi: medRxiv preprint near term peak is not obvious and the peak is attained at around 209,000 daily new cases ( Figure 3A and 3B) ; it would have induced a reduction (approximately 51.9% or 5.4 million cases) in total case counts from April 15 to May 15, 2021, but the 95% CI [-0.5, 9.9] includes the zero value so the decrease is not statistically significant. An end-of-April national lockdown would have been too late for any marked benefit because the second wave outbreak had already run its course (largely aided by state-level lockdowns in Maharashtra and Delhi (59)), as is evident by the slowing growth and eventual decline of observed daily cases after May 6 (14) .

To estimate the number of preventable deaths we multiply the predicted cases under each intervention effect with daily CFR estimates obtained under three different scenarios. Details are presented in Supplementary Section 3. Three daily CFR schedules, based on observed data from Kerala, India, and Maharashtra on exactly the same dates, are applied to daily predicted case counts. For simplicity, we refer to these as low, moderate, and high CFR schedules, respectively. A similar pattern as the case counts can be seen with respect to death counts. Clear (Figure 4C) , if there is sufficient healthcare capacity (i.e., low CFR scenario), we see that a lockdown would have remarkable benefits in terms of reducing death counts. For example, we see that a March 15 lockdown would have resulted in a . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 30, 2021. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 30, 2021. 

By virtually every metric, Wave 2 of the COVID-19 pandemic in India has been more acute and devastating than Wave 1. Owing to the development of COVID-19 vaccines, treatments and care, and the fact that variants of the virus have not yet demonstrated marked increases in lethality, the observed case fatality rate appears to be lower in Wave 2 than in Wave 1 (31) . However, with daily infections growing exponentially, even with a small infection fatality rate, a large number of deaths are bound to occur, partly due to the natural infection related mortality but also due to the collapse of the healthcare system. We see a tragic example of this in India's Wave 2. Despite the early and decisive actions taken in Wave 1, strong interventions were not enforced in Wave 2. Our analysis shows that a large fraction of cases and deaths could potentially be averted with early comprehensive nationwide interventions, with realistic data-driven effect sizes that are derived from lockdowns that took place in India.

We acknowledge there are certain overarching limitations in our work.

First, under-reporting of cases and deaths attributed to COVID-19 is not aptly accounted for across these results. At the time of this report, officials have reported intermittent excess death calculations for selected cities (60-62) and for even fewer states in India, and hence, adjusting for under-reporting remains a challenge (62) . All reports point to a large degree of underreporting, particularly in rural India (48) . Similar considerations apply for infections. While we characterize effects of NPI on reported cases this only captures . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 30, 2021. ; https://doi.org/10.1101/2021.06.23.21259405 doi: medRxiv preprint a small fraction of infections. The more recent serosurveys that are emerging indicate 55% seropositivity in age-groups 0-17 and 63% in adults in urban areas (63) . Second, the lack of disaggregation of cases and deaths attributed to COVID-19 in the data available at the time of this study prohibits any investigation into differences within age-sex strata and identify vulnerable, underserved subgroups. This limits our ability to interpret some of the observations. For example, the apparent overall lower infection fatality rate in Wave 2 can be largely due to younger people getting infected. An age-specific fatality comparison is necessary but could not be performed. Third, our models do not incorporate vaccine roll-out. For further context, during the Wave 2 analysis period, about 3% of India was fully vaccinated and 10% received at least one dose (based on vaccine data available from covid19india.org (14) through May 15, 2021). Since age and occupation was at large the determining factor in vaccine eligibility, accounting for vaccine distribution during this period also requires age-stratified data which is not currently available in India. Fourth, our assessment period for the effect of intervention contains the period April 15-May 15 where many states in fact instituted partial lockdowns. Thus, the comparison with the observed data for this period is not with a no-intervention scenario but with the actual ground reality with a mix of mitigation strategies being compared to the idealized hypothetical lockdown effect. The comparison up to April 15 is clearer to interpret. Finally, we had to restrict our exploration of the epidemiologic properties of B.1.617.2 to the state of Maharashtra as the nationwide temporal distribution of this variant that is available in the GISAID database does not have adequate coverage of all Indian states. Amongst other states with genomic data, we chose Maharashtra not just because it contributes the largest share to India's COVID case and death burden, but . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 30, 2021. ; https://doi.org/10.1101/2021.06.23.21259405 doi: medRxiv preprint also because it instituted the earliest statewide lockdown (April 14, 2021). If we see a high transmissibility despite a statewide NPI, it provides stronger evidence in support of the hypothesis. However, we do acknowledge as shown in Supplementary Figure S8 a,b, growth of SARS-Cov-2 variants in India has very large spatiotemporal heterogeneity between March 1-May 15, 2021 (64, 65) . A detailed future analysis is required to really capture this heterogeneity once reliable genomic, mortality and seroprevalence data is available for the second wave. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 30, 2021. ; https://doi.org/10.1101/2021.06.23.21259405 doi: medRxiv preprint vaccinated is by the end of this year (68) . However, India is a very young population and roughly 40% of the population is in the age group of 0-18 years (69), for whom there are no vaccines available yet. Until there is a sufficient supply of vaccines to inoculate most of the Indian population, non-pharmaceutical public health interventions will be key for India as the country designs its current lockdown exit strategy. Such measures are crucial for curbing deaths, infections, and subsequent viral mutation; out analysis demonstrates that the earlier an intervention takes place, the better -timing matters.

Our findings echo the need for enhanced surveillance efforts, continued attention to the emergence of variants and examination of subsequent consequences for transmissibility, cross-immunity, and vaccine effectiveness. The scenarios we ran for the models with sequencing data show us that the resurgence of the second wave cannot be fully explained by the waning immunity or even the increased transmissibility. The extent of reinfections or immune evasion can only be substantiated after clinical studies of the variants, and we also need better availability of data so that we can move from modelling scenarios to more data-based inferences to provide better understanding of the epidemic landscape in India. We urge further ramping up of the sequencing effort in India to ensure sequencing-based epidemiological analyses are equipped with a sufficiently large, representative sample and to keep pace with, as well as predict, the various evolutionary paths of the virus. Strengthening the surveillance for COVID19 in every district is important for identifying clusters requiring further investigation. Among existent innovative strategies, wastewater analysis has also been shown to be a promising method for crossvalidation of clinical data, for evaluating regional genomic sequencing efforts, and for in turn identifying novel SARS-CoV-2 strains otherwise undetected (70) . Prioritizing . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted June 30, 2021. ; https://doi.org/10.1101/2021.06.23.21259405 doi: medRxiv preprint sequencing of reinfections, as well as vaccine breakthrough infections, is necessary to understand vaccine effectiveness and development needs against emerging Variants of Concern. We hope that the lessons from Wave 2 lead to a bolstering of public health infrastructure, timely and comprehensive collection, and release of data as well as compel policymakers to act more proactively, thereby preparing India to respond to future waves and crises.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted June 30, 2021. ; https://doi.org/10.1101/2021.06.23.21259405 doi: medRxiv preprint this study is based (full author list attached as supplementary material). 

We use data on reported infected cases and COVID-19-attributed deaths through May 31, 2021, for our descriptive analysis (Section 1) and May 15, 2021, for predictive modeling (Sections 2 and 3) from covid19india.org (14) . Population data for India was obtained from Our World in Data (71) . Genomic data was obtained from GISAID (49, 50) . Analysis code are available at on the Center for Precision Health Data Science GitHub page: https://github.com/umich-cphds/covid_india_wave2.

Materials and Methods Sections 1 -3 Table S1 -S9 Fig S1 -S8  Supplementary references 1 -27 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted June 30, 2021. 

First confirmed case of COVID-19 infection in India: A case report

COVID-19 India Timeline. The Wire

WHO Director-General's opening remarks at the media briefing on COVID

Predictions, Role of Interventions and Effects of a Historic National Lockdown in India's Response to the the COVID-19 Pandemic: Data Science Call to Arms

Centre extends nationwide lockdown till May 31, new guidelines issued

Modi Orders 3-Week Total Lockdown for All 1.3 Billion Indians

India to remain closed till 3 May, economy to open up gradually in lockdown 2.0. mint (2020)

ANI, India scales up testing capacity from 1 in January to over 7.7 crore in October: MoHFW. Econ. Times (2020)

Updates on COVID-19, (available at pib.gov.in/Pressreleaseshare.aspx?PRID=1615405)

India can expect about 4,92,380 total deaths by December 1, says US study. @businessline

Coronavirus in India: Latest Map and Case Count

Incorporating false negative tests in epidemiological models for SARS-CoV-2 transmission and reconciling with seroprevalence estimates

Prevalence of SARS-CoV-2 in Karnataka

Population-Based Serosurvey for Severe Acute Respiratory Syndrome Coronavirus 2 Transmission

Delhi's 5th sero survey: Over 56% people have antibodies against Covid-19

Community prevalence of antibodies to SARS-CoV-2 and correlates of protective immunity in an Indian metropolitan city

ICMR sero survey: One in five Indians exposed to Covid-19

India Approves Oxford-AstraZeneca Covid-19 Vaccine and 1 Other

COVID-19 Vaccines Operational Guidelines

A New Variant May Be The Cause Of India's COVID-19 Surge

Maharashtra imposes 'mini lockdown' amid rising coronavirus cases

India's COVID-19 emergency. The Lancet

Dexamethasone in Hospitalized Patients with Covid-19

How bad is the second wave? BBC News (2021)

Covid-19 in India: Why second coronavirus wave is devastating

The scale of Gujarat's mortality crisis

Estimating the wave 1 and wave 2 infection fatality rates from SARS-CoV-2 in India

The Indian SARS-CoV

Genomic characterization and Epidemiology of an emerging SARS-CoV-2 variant in Delhi

Emergence of a new SARS-CoV-2 variant from GR clade with a novel S glycoprotein mutation V1230L in West Bengal

COVID-19 time-varying reproduction numbers worldwide: an empirical analysis of mandatory and voluntary social distancing

Genome Sequencing by INSACOG shows variants of concern and a Novel variant in India, (available at pib.gov.in/Pressreleaseshare.aspx?PRID=1707177)

Pan-India novel coronavirus SARS-CoV-2 genomics and global diversity analysis in spike protein

Convergent evolution of SARS-CoV-2 spike mutations, L452R, E484Q and P681R, in the second wave of COVID-19 in Maharashtra

Maharashtra: double mutant found in 61% samples tested

Importation, circulation, and emergence of variants of SARS-CoV-2 in the South Indian state of Karnataka

per cent of Punjab COVID-19 samples sent for genome sequencing show new UK variant

Coronavirus | New virus lineage found in West Bengal. The Hindu (2021)

Why India's massive COVID-19 surge is breaking tragic records

India's devastating second wave: 'It is much worse this time

Representative Estimates of COVID-19 Infection Fatality Rates from Three Locations in India

Bihar revises Covid toll, deaths jump to over 9

As Covid-19 Devastates India

Data, disease and diplomacy: GISAID's innovative contribution to global health: Data, Disease and Diplomacy

GISAID: Global initiative on sharing all influenza data -from vision to reality

The Center for Viral Systems Biology, outbreak.info. outbreak.info (2020)

SARS-CoV-2 Variants of Interest and Concern naming scheme conducive for global discourse

Neutralising antibody activity against SARS-CoV-2 VOCs B.1.617.2 and B.1.351 by BNT162b2 vaccination

Risk assessment for SARS-CoV-2 variant: Delta (VOC-21 APR-02, B.1.617.2)

Imperial College COVID-19 Response Team

Estimating the effects of non-pharmaceutical interventions on COVID-19 in Europe

Prevalence of SARS-CoV-2 infection in India: Findings from the national serosurvey

Covid-19 second wave: Here's a list of states that have imposed full lockdown

Estimating COVID-19 infection fatality rate in Mumbai during 2020

Reconstructing the COVID-19 epidemic in Delhi, India: infection attack rate and reporting of deaths

Estimating COVID-19 Related Mortality in India: An Epidemiological Challenge with Insufficient Data

Serological prevalence of SARS-CoV-2 antibody among children and young age (between age 2-17 years) group in India: An interim result from a large multi-centric populationbased seroepidemiological study

Nextstrain: real-time tracking of pathogen evolution

SARS-CoV-2 B.1.617 emergence and sensitivity to vaccineelicited antibodies

SARS-CoV-2 variant B.1.617 is resistant to Bamlanivimab and evades antibodies induced by infection and vaccination

37 Ist, By end of 2021, India will be in position to vaccinate all adults: Health Minister Harsh Vardhan. India Today

High throughput sequencing based detection of SARS-CoV-2 prevailing in wastewater of Pune

Coronavirus Pandemic (COVID-19)

Maximum # Daily New Cases

4x, 236% (13,560 on June 22 to 45,601 on July 22) 8.2x, 722% (15,353 on March 8 to 126,276 on April 7) Biggest 30-day relative increase in deaths 3.6x, 262% (312 on June 22 to 1,130 on July 22) 13.7x, 1270% (266 on March

) total number of predicted cases since the start of lockdown in the second row (with 95% CI), (3) the number of cases averted (relative to observed) since the start of lockdown in the third row

) total number of predicted deaths since the start of lockdown in the second row (with 95% CI), (3) the number of deaths averted (relative to observed) since the start of lockdown in the third row (with 95% CI), and (4) the relative reduction in deaths (as a percent)

The authors thank Lili Wang for her advisement regarding the use of her eSIR R package. The authors also thank authors from the originating laboratories and submitting laboratories for their collection, generation, and proliferation of the GISAID data on which