The relative infectivity of the brand new UK variant of SARS-CoV-2 – Watts Up With That?

The relative infectivity of the new UK variant of SARS-CoV-2 – Watts Up With That?

Reposted by Dr. Judith Currys Climate Etc.

Posted on December 29th, 2020 by niclewis

By Nic Lewis

Important points

  • A new variant, B.1.1.7, of the SARS-CoV-2 virus has recently spread rapidly in England
  • The health department's best estimate for the weekly growth rate advantage of B.1.1.7 is 1.51x
  • You are incorrectly converting this to a reproduction ratio of 1.47; A reasonable conversion gives a ratio of 1.25
  • Confident claims by the UK government's scientific advisors that the higher growth of B.1.1.7 was due to increased portability are false. it could be partly due to other factors entirely
  • 1.1.7 has shown no greater growth rate advantage than two earlier variants, both of which are now assumed to have no greater transferability than previously existing variants
  • There is little evidence that B.1.1.7 is more virulent or likely to be resistant to existing vaccines


The apparently rapid growth in England of a new variant of the SARS-CoV-2 virus causing COVID-19 has sparked dire warnings from UK government advisors. Their advice only indicated that the new variant was more communicable (more infectious) and not more virulent (causing more serious diseases). Nonetheless, this led to swift (many would say panic) action first by the British government and then by the governments of many other countries. The UK government has further restricted people's freedom to mingle and move around while other countries have banned travelers from the UK. Millions of people in Great Britain had to cancel their Christmas holiday plans at very short notice and then continue to restrict their freedoms. In this article I examine how justified the advice that led to these harmful government actions was.

The new strain B.1.1.7 and its distribution in Great Britain

The UK government agency Public Health England (PHE) named the new variant VUI-202012/01 and now VUC-202012/01, but I will refer to them by the scientific name of their lineage B.1.1.7 (Rambaut et al.) ( 1) or simply as "the new variant". The line comprises 8 amino acid changes (6 mutations and 2 deletions) (2) in the gene for the important spike protein and 9 amino acid changes (3) in genes for other proteins. The line has sometimes just been referred to by the name of the most famous mutation it possesses, N501Y. However, this should be avoided as there are other variants that also have this mutation.

Rambaut et al. have this to say about the new line:

The B.1.1.7 line shows a greater number of viral genetic changes than usual. The accumulation of 14 line-specific amino acid substitutions prior to their detection is unprecedented in global virus genome data for the COVID-19 pandemic.

They also identify three of the mutations in particular (including N501Y) that are suspected of having potential biological effects.

B.1.1.7 was first detected in SARS-CoV-2, which was sequenced from a sample taken on September 20, 2020 in the south-east of England. Since then, the number of cases has grown rapidly and spread to other locations. The UK sequences many more SARS-CoV-2 genomes than any other country and more than the rest of Europe combined. The fact that B.1.17 was first detected in the United Kingdom does not necessarily mean that it originated there. The line has also been discovered in several other countries and may be widespread.

The growth of the B.1.1.7 line in the UK can be followed from sequencing data uploaded to GISAID. I used the COVID-CG processing tool (4) to select each day's sequences with all eight B1.1.7 spike gene mutations. (5) Since the daily data was noisy and only a few sequences were dated after December 12th, 2020, I took 7 day moving averages centered until December 9th. Figure 1 shows the resulting proportion of all UK sequences that have been represented by the B.1.1.7 line since it first appeared. It should be noted that the proportion of non-B.1.1.7 sequences represented by the areas where B.1.1.7 first gained importance may have increased over time, which leads to growth that shows how fast it has grown in individual areas or in the UK as a whole.

Illustration 1. The proportion of all SARS-CoV-2 genomes sequenced in Great Britain that is formed by the B.1.1.7 line

The higher growth rate of B.1.1.7

A PHE report published on December 21, 2020 (6) contains epidemiological evidence for the growth rate of B.1.1.7 versus non-B.1.1.7 lines. By using a proxy marker for B.1.1.7 (7), they were able to use data from a significant part of the British "Pillar 2" test program. This provided a much larger dataset than the sequenced SARS-CoV-2 genomes and allowed weekly data to be stratified for each of the 42 NHS-STP areas. Figure 2 reproduces Figure 1 of the PHE document, in which the multiplicative benefit of the weekly growth rates of B.1.1.7 cases (the ratio of B.1.1.7 to non-B.1.1.7 week t + 1 Cases divided by) is shown week t cases). The x-axis is negative for the B.1.1.7 proxy, S gene test. The week indicated is the base week, so the yellow dots reflect the ratio of week 49 cases (week ending December 5) to week 48 cases.

Figure 2. Empirical data analysis of the multiplicative advantage for weekly growth rates. Each point represents the ratio of weekly growth rates between B.1.17 (VOC) and non-B.1.1.7 for an STP area and a week from NHS England, based on Pillar 2 data in Figure S1 of the PHE report. Colors and shapes differentiate calendar weeks. Numbers above 1 show a multiplicative advantage. The blue line represents the mean value for a certain frequency and the gray lines the 95% envelope. The scatter at low frequencies largely reflects the statistical noise due to low counts.

If the new variant makes up a small proportion of the total cases (e.g. less than ~ 25%), the proxy used is less satisfactory and there is also a lot of variation. Even so, the variant's proxy-based mean multiplicative advantage (ratio) in weekly growth is remarkably independent of its relative prevalence. This supports the PHE methodology, although week 48 data suggests that the multiplicative benefit might decrease once the variant accounts for more than ~ 25% of the total cases. From this data, PHE calculates a mean multiplicative advantage for weekly growth of 1.51 for B.1.1.7. Assuming a fixed generation interval of 6.5 days, they convert this into a multiplicative advantage of the reproduction number (Rt) of 1.47 for B.1.1.7 over other variants.

PHE also estimated the effect of B.1.1.7 on Rt using genomic data (sequencing data) for the same areas and weeks. They estimated an additive effect on Rt of 0.57 or 0.74 if the effect was allowed to vary between areas. PHE also estimated the effect on Rt using the S-gene proxy data from the PCR assay adjusted for specificity (which is bad when the negative level of the S-gene is low). Their estimates of the additive effect on Rt using this data were 0.52 or 0.60 if the effect was allowed to vary between areas. Using a Bayesian regression model, their estimate of the effect was 0.56. However, since a biological difference in infectivity would be expected to cause a multiplicative effect on Rt and Rt varied over the analysis period, an estimated additive effect on Rt is less useful and also less accurate than a multiplicative estimate. In addition, all of these estimates include more complicated statistical models, other assumptions, and estimates of other variables. So I prefer their estimated multiplicative advantage of 1.51 (for weekly growth before moving to the RT scale), which is derived directly from the underlying data. This corresponds to a logarithmic daily growth rate advantage of 0.059.

Further information on the faster growth of B.1.1.7

On December 18, 2020, a meeting of the NERVTAG (8) committee took place, which advises the government on the threat of new and emerging respiratory viruses. The protocols (9) refer to an estimate from genomic data of a 71% higher growth rate than other variants; None of the documents examined by the committee contained such an estimate. The minutes of a subsequent meeting on December 21, 2020 (10) show that this was one of several undocumented estimates by NERVTAG member Professor Neil Ferguson of Imperial College. An alternative regression estimate that he apparently submitted showed that line B.1.1.7 had a value 0.39 Rt higher than non-variant lines from early November to early December. This is presumably an additive effect estimate and is well below the estimates of PHE, which use largely the same method. Two other estimates given in Professor Ferguson's minutes actually appear to be slightly misrepresented versions of the PHE estimate of a multiplicative Rt advantage of 1.47 for B.1.1.7.

The minutes of another NERVTAG meeting on December 21 also mention an estimate by the London School of Hygiene and Tropical Medicine that B.1.1.7 was 56% more transferable (a multiplicative advantage of 1.56 in RT value) . This estimate is documented in a preprint (Davies et al. (11)). The authors use a subjective Bayesian method to fit a highly complex model with many probabilistic parameters, some of which are fixed and some of which are estimated. This is far from a robust and potentially inherently biased method of estimating the relative transfer rate. Additionally, they seem to use less informative data broken down geographically only by the 7 NHS regions, not (as PHE uses) by the 42 NHS STP areas. Using less informative data implies that even if they had applied a robust method, their estimates would inherently be less reliable than the PHE estimate. The uncertainty ranges of their estimates – including a confidence interval of 99% + from 1.49x to 1.57x for the Southeast region (12) – seem quite unrealistically narrow given the significant uncertainties. This raises further doubts about the realism of their estimates.

Finally, the NERVTAG meeting on December 21 also took into account a phylogenetic estimate by the University of Edinburgh (Andrew Rambaut) based on genetic sequences from Kent and London, according to which Rt was 1.57 or 1.72, depending on the time window used. Since no comparative Rt estimate for non-B.1.1.7 variants is mentioned in the meeting minutes, it is not possible to derive an estimate of the relative transmission rate of the new variant from this.

I conclude that the other evidence considered by NERVTAG is less robust and less useful than the PHE estimate of a multiplicative advantage in weekly growth of 1.51.

Why the faster growth of B.1.1.7 does not have to be due to an increase Transferability

While the evidence that the B.1.1.7 line grew faster than other lines in England in the two months to early / mid-December appears robust, the biological properties of a virus cannot be inferred from limited epidemiological data. The apparently rapid spread of this new variant could be due more to founder effects and super-spreader events than to or additionally to increased transferability (higher infectivity).

In this regard, it is instructive to look at two other lines / variants which also had an exceptionally rapid growth phase and in one case became completely dominant in most countries.

The G clade: spike gene mutation D614G

The D614G mutation appeared early in the epidemic that hit Europe in February, and the G614 variant undoubtedly spread faster than D614 in most places. In a very large number of countries, areas and cities, it went from a minority of infections to the dominant variant within a month. Since July 2020, it has had a share of almost 100% of new infections in most countries and on all continents.

Given that D614G is becoming and remaining so dominant, it is not surprising that an August release (Korber et al. (13)) argued that the D614G mutation increased transmissibility, citing several pieces of evidence:

  • the consistency of the increase (in the frequency of G614) across geographic regions.
  • The D614 form did not persist in many places where the G614 form was introduced into the ongoing, well-established D614 epidemics, as would be expected if the two forms were equally likely to spread.
  • The rise in G614 frequency often continued well after national stay-at-home orders (lockdowns) were introduced, which were likely to have significantly reduced serial overseeding by travelers.

Furthermore epidemiological evidence, tThe authors also found that increased transmissibility of G614 is consistent with other studies suggesting associations with increased infectivity in vitro (14) (15), and with their own finding an association with higher viral loads in vivo. In addition, another publication (Li, Q et al. (16)) reported higher antigenicity for G614.

Most of Korber's arguments are also applicable to evidence suggesting that B.1.1.7 may be more transferable. A more recent publication in Nature (van Dorp et al. (17)), however, found “no evidence of significantly more transmittable lines of SARS-CoV-2 due to recurring mutations”, including D614G (B.1.1.7) until the end of the study period identified). This shows that conclusions about its relative transferability from epidemiological and indirect biological evidence can turn out to be false even after the dominance of a new strain.

The 20A.EU1 variant: Spike mutation A222V

The 20A.EU1 variant, in which the spike gene mutation A222V is involved, appeared in Spain in early summer 2020. It quickly spread to other European countries, where it typically grew faster than non-20A.EU1 variants. Figure 3 shows the log daily growth rate of sequences with the A222V mutation in the UK compared to those without it over the two months to mid-September. During this period, the ratio of A222V to new sequences without A222V grew from below 0.02 to 0.67. The log mean daily growth rate was 0.061 – a weekly multiplicative benefit of 1.53 – with no substantial trend. This multiplicative advantage is effectively the same as PHE's estimate of 1.51 for B.1.1.7 using data from mid-October to mid-December.

Figure 3. The log daily rate of growth of the 7-day moving average of new sequences with the A222V mutation in the UK compared to those without this mutation for the two months ended September 12, 2020.

In the autumn, however, the relative frequency of new A222V sequences no longer increased in a number of countries without achieving complete and continuous dominance as with D614G (Figure 4). In Great Britain, the A20.EU1 variant had reached around 70% of all new sequences by the end of October, but has since declined relatively often, as was the case in Belgium, Germany and Switzerland.

Figure 4. The proportion of weekly new SARS-CoV-2 sequences in 2020 in ten European countries with the A222V mutation (which means that it is the A20.EU1 variant)

Regardless of the rapid growth of the 20A.EU1 variant in many European countries in summer and / or autumn, a preprint paper from November 2020 on 20A.EU1 came to the conclusion: “We do not find any evidence of an increased transferability of this variant, but show How rising is the incidence in Spain, the resumption of travel across Europe and the lack of effective screening and containment could explain the success of the variant. "(18)

Overestimation of the impact of a possible increased transferability on Rt

Assuming that the previously higher growth rate of the B.1.1.7 line were all due to higher transferability. What effect would this have on the current reproduction number Rt? That depends on what Rt is and the mean generation interval and its probability distribution. The longer the generation interval, the higher the Rt to generate a given growth rate. In the December 21st PHE release, their estimate of the multiplicative benefit of the weekly growth rate (of 1.51) is converted to a multiplicative benefit in Rt of 1.47 by assuming a fixed generation interval of 6.5 days: 1, 47 = 1.51 ^ (6.5 / 7).

However, the PHE conversion formula is not justified for two reasons:

  • the generation interval is not specified; and
  • The most recent estimates of the mean generation interval are well below 6.5 days.

Most generation interval estimates (the amount of time one person is infected to infect another person) are actually serial interval estimates (the time from when one person has symptoms to when they have symptoms when they have infected them) ) Time of infection cannot be observed. The generation interval can be validly estimated by combining probabilistic estimates of the serial interval and the incubation time (from infection to onset of symptoms). However, it is unsatisfactory to treat probabilistic serial interval estimation simply as a representation of the generation interval distribution, as is typically done.

PHE do not provide a source for assuming a generation interval of 6.5 days, but they could follow the Imperial College team who (in Flaxman et al. (19)) used and reported a generation interval with a mean of 6.5 days that it was estimated by Bi et al. (20). In fact, Bi et al. estimated the serial interval, not the generation interval, and fitted a gamma probability distribution with a mean of 6.3 days. However, their data included cases where the infecting individual did not isolate himself from others until long after symptoms appeared. Bi et al. found that when the infected individual was isolated less than three days after symptoms appeared, the average serial interval was only 3.6 days, which is usually the case in the UK.

Knight and Mishra (2020) (21) show that the serial interval, in order to avoid an overestimation, has to be adapted to a probability distribution which, in contrast to the one described by Bi et al. The gamma distribution used allows negative values ​​(which are observed in a non-negligible proportion of cases). They take into account a number of estimates examined in a review of incubation time and serial interval and choose the only serial interval estimate based on a negatively feasible probability distribution with a large sample size (almost ten times the size in Bi et al.) And the incubation time estimate based on the largest sample. The resulting generation interval estimate has a mean of 3.99 days and a standard deviation of 2.96 days. (22)

Davies et al. say that their complex Bayesian model, which estimated a multiplicative advantage of the Rt value of 1.56 using a fairly long generation interval, fitted less well when they used a shorter interval. However, it is likely that the main reason they get a poor fit to the relative growth of the new variant during lockdown with a shorter generation interval is that their model greatly overestimates the effect of the November lockdown. (23)

Using the estimated distribution of Knight and Mishra for the generation interval and the correct conversion formula (24), the PHE estimate of the multiplicative advantage of the B.1.1.7 line in the weekly growth rate of 1.51 equals a multiplicative advantage in Rt of 1 , 25. (25) This is only about half of the PHE estimate of the multiplicative advantage of 1.47. (26) This implies that less extensive and serious measures would be required to prevent exponential growth of infections if B.1.1.7 occurs than implied The PHE estimate of a multiplicative advantage of 1.47 in Rt, even if the observed multiplicative advantage in the weekly growth rate from B.1.1.7 to mid-December is entirely due to the fact that it is more transferable.

The new South African variant

A new SARS-CoV-2 line has recently emerged in South Africa, which also includes an N501Y spike gene mutation and a number of other mutations (which differ from those in B.1.1.7), as described by Tegally et al. (27) who call it 501Y.V2. They say genomic data showing the rapid shifting of other lineages suggests that that lineage might be linked to increased portability. However, the limited evidence available so far is insufficient to justify the alarming remarks made by the UK Health Secretary: “This new variant is extremely worrying as it is even more transmissible and appears to be more mutated than the new variant discovered in the EU Great Britain ". (28) As a professor of molecular virology at the University of Nottingham noted, the mutation in question has been previously observed. We have no idea if it affects virus transmission and we should avoid panic at this point. (29)


The examples G-Klade and 20A.EU1 illustrate that there is apparently strong epidemiological evidence of a higher growth rate of a new variant over a considerable period of time, even when this leads to what appears to be permanent dominance, although accompanied by evidence suggesting that the variant is associated with a higher viral load does not allow a valid conclusion that the variant is more transmissible than existing variants. Such evidence may indicate greater portability, but is not reliable.

Nonetheless, the NERVTAG Committee on December 18th came to the conclusion with moderate confidence that the new variant "has a significant increase in portability compared to other variants" and continued at its meeting on December 21st and expressed great confidence, that "B.1.1.7 can spread faster than other SARS-CoV-2 virus variants currently in circulation in the UK". Although it was admitted that the underlying cause of faster spread was unclear, the causal factors they suggested related solely to higher transmissibility. Given that NERVTAG's confident conclusions are not justified by the facts, it is not surprising that a number of experts in the UK and elsewhere have denied them (30) or have expressed opposing views (31) (32) (33 ). The mainstream media falsely report that B.1.1.7 has been shown to have significantly increased transferability.

I have argued that PHE's estimate of a multiplicative weekly growth benefit of 1.51 for B.1.1.7 is more robust and accurate than the other estimates available for several reasons. I have shown that even if the higher growth of B.1.1.7 so far were solely due to increased transferability, this would correspond to a multiplicative advantage in Rt of only 1.25, half the advantage of PHE below Use of an inappropriate calculated conversion formula.

So far there is no evidence that B.1.1.7 is more virulent than existing strains and is also not resistant to the vaccines developed. Expert opinion seems to be that neither is likely to be the case. (34)

These results suggest that B.1.1.7 does not currently represent a serious increase in the SARS-CoV-2 threat, even in the worst case that its higher observed growth rate is solely due to increased transferability. At best, it will turn out that its higher growth rate is in no way due to increased transferability, as now seems to be the case with the G-clade and the 20A.EU1 variant.

Accordingly, it is difficult to see that taking drastic measures to slow down transmission, further reduce economic activity, social activity and human freedom is justified by the current evidence on the emerging B.1.1.7 line.

However, as soon as more evidence becomes available, it could reinforce or weaken the case that the emergence of B.1.1.7 is a serious development. It is important that the UK authorities publish all available data on cases of the new strain on a daily basis and at the local authority or detailed level, as indicated by the "S-Gen negative" proxy and any other method. Currently, they do not keep this information public, which makes it impossible for independent researchers to evaluate in time and, if necessary, to question the possibly incorrect conclusions. In addition, it is highly desirable that the government or its advisors not review a SARS-CoV-2 or COVID-19-related report or study below, unless accompanied by a link that lists all of the data used Are available.

Nicholas Lewis December 29, 2020

(1) Rambaut, A et al .: Preliminary genomic characterization of an emergent SARS-CoV-2 line in Great Britain, defined by a new set of spike mutations. COVID-19 Genomics Consortium UK, ARTIC Network December 19, 2020. -defined -by-a-new-set-of-spike-mutations / 563

(2) N501Y, A570D, P681H, T716I, S982A and D1118H mutations plus HV69-70 and Y144 deletions

(3) T1001I, A1708D and I2230T mutations plus SGF3675-3677 deletion in the ORF1ab gene; R52I and Y73C mutations plus Q27stop codon in the Orf8 gene; D3L and S235F in the N gene. There are also 6 synonymous (non-amino acid changing) mutations: 5 in ORF1ab (C913T, C5986T, C14676T, C15279T, C16176T) and 1 in the M gene (T26801C).

(4) Data downloaded December 26, 2020.

(5) Only the total number of sequences of each day with each mutation is available via COVID-CG, but the number of each of the eight spike mutations that occur each day (r> 0.999, except 0.991 for the HV69-70 deletion, which sometimes occurs occurs with other strains) meaning they have extremely high concurrent occurrences. I took the minimum number for the eight spike mutations as the count for B.1.1.7 sequences each day. Including non-spike mutation data appears unnecessary. The agreement of all B.1.1.7 spike mutations with all ORF1ab B.1.1.7 mutations is almost perfect, with the exception of A1708D, which is approximately 1% of the cases in which all 11 other spike and ORF1ab mutations are present are, does not appear to be present.

(6) Public Health England: Investigation of the novel SARS-COV-2 variant – variant of concern 202012/01.

(7) S-gene negative, N-, and ORF1ab-positive TaqPath PCR test result.

(8) NERVTAG: Advisory Group on New and Emerging Threats from Respiratory Viruses

(9) NERVTAG COVID-19 VUI communication 18122020_final.pdf, available at

(10) NERVTAG COVID-19 VOC communication 21122020 final.pdf, available at

(11) Davies, NG et al .: Estimated transferability and severity of the novel SARS-CoV-2 variant of concern 202012/01 in England. Zentrum für mathematische Modellierung von Infektionskrankheiten, Londoner Schule für Hygiene und Tropenmedizin, aktualisiert am 23. Dezember 2020.

(12) Abbildung 1A von Davies et al., Ganz rechts.

(13) Korber, B. et al. Verfolgung von Änderungen der SARS-CoV-2-Spitze: Hinweise darauf, dass D614G die Infektiosität des COVID-19-Virus erhöht. Cell 182, 812.e19–827.e19 (2020)., 20. August 2020

(14) Zhang, L. et al. Die D614G-Mutation im SARS-CoV-2-Spike-Protein reduziert die S1-Abgabe und erhöht die Infektiosität. Preprint unter, 12. Juni 2020

(15) Yurkovetskiy, L. et al. Struktur- und Funktionsanalyse der D614G SARSCoV-2-Spike-Protein-Variante. Cell 183, 739.e8–751.e8, Oktober 2020

(16) Li, Q. et al. Der Einfluss von Mutationen in der SARS-CoV-2-Spitze auf die virale Infektiosität und Antigenität. Cell 182, 1284.e9–1294.e, September 2020

(17) van Dorp, L et al., Keine Hinweise auf eine erhöhte Übertragbarkeit aufgrund wiederkehrender Mutationen in SARS-CoV-2. Nature, November 2020.

(18) Hodcroft, BH et al.: Entstehung und Verbreitung einer SARS-CoV-2-Variante in Europa im Sommer 2020. medRxiv 27. November 2020

(19) S. Flaxman, S. Mishra, A. Gandy et al. Abschätzung der Auswirkungen nichtpharmazeutischer Interventionen auf COVID-19 in Europa. Nature 584, 257–261 (2020).

(20) Bi, Q. et al. Epidemiologie und Übertragung von COVID-19 in Shenzhen, China: Analyse von 391 Fällen und 1.286 ihrer engen Kontakte. medRxiv (2020)

(21) Knight, J. und Mishra, S.: Schätzung der effektiven Reproduktionszahl unter Verwendung der Generierungszeit gegenüber dem seriellen Intervall mit Anwendung auf COVID-19 im Großraum Toronto, Kanada. Modellierung von Infektionskrankheiten 5 (2020) 889e896, November 2020.

(22) Knight und Mishra passen ihre Generierungsintervallschätzung unter Verwendung einer Gammaverteilung an. Im Gegensatz zum seriellen Intervall kann das Generierungsintervall nicht negativ sein, daher ist hier eine Gammaverteilung geeignet.

(23) Sie sagen, dass die schlechte Anpassung mit einem kürzeren Generationsintervall darauf zurückzuführen ist, dass vorausgesagt wurde, dass die relative Häufigkeit des neuen Stammes während der Sperrung im November abgenommen haben sollte. Wie sie schreiben: "Wenn Rt <1 für beide Varianten ist, ist eine kürzere Generierungszeit ein selektiver Nachteil, da Infektionen mit dieser Variante schneller abnehmen als mit einer Variante mit demselben Rt, die jedoch auf einer längeren Zeitskala übertragen wird." Dies gilt jedoch nur, wenn Rt während der Sperrung unter 1 lag, während ihre 1E zeigt, dass Rt während der Sperrung in Wirklichkeit bei oder geringfügig über 1 blieb. Dies steht im Einklang mit den Mobilitätsdaten von Davies et al. Fig. 1C, die einen geringen Unterschied zwischen unmittelbar vor dem Start und dem Ende der Sperrung zeigen. Ein Gesamt-Rt von 1 impliziert, dass die Infektionen mit der übertragbareren Variante in der relativen Häufigkeit zunehmen, wie sie aufgetreten sind, nicht abnehmen. Die Komplexität ihres Modells bedeutet auch, dass die schlechte Anpassung teilweise oder vollständig auf andere Ursachen zurückzuführen sein kann, beispielsweise auf ein langes Generierungsintervall, das die Fehlschätzung eines anderen Parameters kompensiert, oder auf die besondere Art und Weise, in der sie ein verkürztes Generierungsintervall darstellen.

(24) Wallinga, J. & Lipsitch, M. (2007). Wie Generationsintervalle die Beziehung zwischen Wachstumsraten und Reproduktionszahlen beeinflussen. Verfahren der Royal Society B: Biological Sciences, 274 (1609), 599-604. Verwendung ihrer Gleichung 2.9 in Verbindung mit der Funktion zur Erzeugung des Gammaverteilungsmoments.

(25) Die Schätzung des multiplikativen Vorteils geht davon aus, dass Rt für die anderen Stämme 1,0 beträgt; Der abgeleitete multiplikative Vorteil ist eine langsam abnehmende Funktion von Rt für die anderen Stämme.

(26) The same is approximately true throughout PHE’s 95% confidence interval of for the Rt ratio of 1.34–1.59, which when converted in the same way corresponds to an Rt ratio range of 1.19–1.30.

(27) Tegally, Houriiyah, et al. “Emergence and rapid spread of a new severe acute respiratory syndrome-related coronavirus 2 (SARS-CoV-2) lineage with multiple spike mutations in South Africa.” medRxiv 22 December 2020.

(28) As quoted in the Guardian, 23 December 2020.

(29) Professor John Ball, as quoted at

(30) Vincent Racaniello, Professor of Immunology, Columbia University. Extensive detailed comments, including that “none of the isolates so far have proven implications for human transmission or pathogenesis, including the latest variant isolated from the UK.” and, concerning the NERVTAG 21 December meeting: “You can’t use epidemiological data to prove a biological effect of a amino acid change in a virus; you have to do experiments to do that. And that’s what they’re doing here. They say, there is an increase in the transmissibility. It must be because of the variant. Well, obviously that’s a flawed argument. That’s not how we do science.” ; transcript at

(31) Dr Julian Tang, Honorary Associate Professor/Clinical Virologist, University of Leicester, said:
“The spread of this new virus variant could be due to many factors.  As we saw with the earlier D614G variant – just higher viral loads in clinical diagnostic swabs or in cell culture may not necessarily translate to a more transmissible virus at the population level.
“A higher genomic growth rate in the samples sequenced, may not necessarily mean higher transmissibility, e.g. if there was a rave of several thousand people where this variant was introduced and infected many people mostly in that rave, this may seem very high compared to a lower background of non-variant virus, e.g. in an otherwise prevailing national lockdown.”

(32) Professor Vineet Menachery, University of Texas Medical Branch, said: “So this isn’t the first time we’ve seen variants emerge quickly or begin to dominate the population of viruses that we’re seeing. And so I’m not particularly worried at this moment. There is evidence that it is maybe slightly more transmissible, but we’re not at this point knowing enough about it to really be scared in the sense that it’s a different order of magnitude, that it’s going to be a major threat moving forward.”

(33) Dr Nusrat Homaira, Respiratory Epidemiologist, UNSW Sydney. “There is modest evidence suggesting that this new variant of Sars-CoV-2 is more transmissible, and is speculated to be the reason for the recent increase in the number of Covid-19 cases in London, South East, and East of England regions.”

(34) Dr Julian Tang, Honorary Associate Professor/Clinical Virologist, University of Leicester, said: “We are not seeing any increased virulence (clinical severity) or any gross changes in the S (spike protein) that will reduce vaccine effectiveness – so far.”

Originally posted here, where a pdf copy is also available

Like this:

To like Loading…