sign in sign up
<<
Home , Arteriviridae, Caliciviridae, Coronaviridae, Flaviviridae, Hantaviridae, Hepadnaviridae

Supplementary Information for

Viral zoonotic risk is homogenous among taxonomic orders of mammalian and avian reservoir hosts

Nardus Mollentze and Daniel G. Streicker

E-mail: [email protected] / [email protected]

This PDF file includes: Figs. S1 to S7 SI References

Other supplementary materials for this manuscript include the following: Data and all code (available at https://doi.org/10.5281/zenodo.3516613)

www.pnas.org/cgi/doi/10.1073/pnas.1919176117 Nardus Mollentze and Daniel G. Streicker 1 of9 Passeriformes Perissodactyla Carnivora Galliformes Anseriformes

Tog Ort Ort Ort Rha Fla Reo Hepe Phe Picornaviridae Nai Cor Tog

Paramyxoviridae Diprotodontia Togaviridae Flaviviridae Hepe Fla Tog Cetartiodactyla Are Caliciviridae Nai Reo Chiroptera Peribunyaviridae Rhabdoviridae Togaviridae Rhabdoviridae Coronaviridae Flaviviridae Cal Reo

Paramyxoviridae Flaviviridae Picornaviridae Poxviridae

Herpesviridae Flaviviridae Poxviridae Reoviridae Togaviridae Rodentia Hepa Per Pol Arenaviridae Retroviridae Picornaviridae

Proportion zoonotic 0.00 0.25 0.50 0.75 1.00

Fig. S1. Number of zoonotic linked to key mammalian and avian reservoir orders. Each rectangle represents a reservoir – family combination, with its size corresponding to the number of zoonotic virus definitively linked to that reservoir and its colour indicating the proportion of the total number of viruses in that reservoir–family combination that this represents. Ade = Adenoviridae, Are = Arenaviridae, Art = , Asf = Asfarviridae, Ast = Astroviridae, Cal = Caliciviridae, Cor = Coronaviridae, Fla = Flaviviridae, Hepa = , Hepe = , Nai = , Ort = , Pap = , Per = Peribunyaviridae, Phe = , Pic = Picornaviridae, Pne = , Pol = , = Poxviridae, Reo = Reoviridae, Rha = Rhabdoviridae, Tob = , Tog = Togaviridae.

2 of9 Nardus Mollentze and Daniel G. Streicker Adenoviridae Arenaviridae Arteriviridae

Rodentia Rodentia Primates Rodentia Perissodactyla Primates Diprotodontia Perissodactyla Cetartiodactyla Chiroptera Carnivora Cetartiodactyla

Asfarviridae Astroviridae Caliciviridae

Cetartiodactyla Rodentia Carnivora Lagomorpha Cetartiodactyla Galliformes Cetartiodactyla Anseriformes Carnivora

Circoviridae Coronaviridae Filoviridae

Rodentia Perissodactyla Chiroptera Cetartiodactyla Chiroptera Cetartiodactyla Carnivora Galliformes

Flaviviridae Hantaviridae Hepadnaviridae Rodentia Primates Perissodactyla Rodentia Diprotodontia Chiroptera Rodentia Cetartiodactyla Primates Passeriformes Galliformes

Hepeviridae Nairoviridae

Rodentia Rodentia Rodentia Chiroptera Primates Perissodactyla Cetartiodactyla Chiroptera Diprotodontia Carnivora Cetartiodactyla Cetartiodactyla Galliformes Carnivora

Orthomyxoviridae Papillomaviridae Paramyxoviridae

Rodentia Rodentia Cetartiodactyla Primates Chiroptera Perissodactyla Cetartiodactyla Carnivora Lagomorpha Carnivora Cetartiodactyla Galliformes Anseriformes Carnivora Anseriformes

Parvoviridae Peribunyaviridae Phenuiviridae

Rodentia Rodentia Primates Primates Lagomorpha Cetartiodactyla Cetartiodactyla Cetartiodactyla Carnivora Passeriformes

Picornaviridae Pneumoviridae Polyomaviridae Rodentia Primates Rodentia Rodentia Perissodactyla Chiroptera Cetartiodactyla Cetartiodactyla Primates Carnivora Passeriformes Galliformes Galliformes Cetartiodactyla Anseriformes

Poxviridae Reoviridae Retroviridae

Rodentia Rodentia Rodentia Primates Primates Perissodactyla Primates Lagomorpha Diprotodontia Perissodactyla Cetartiodactyla Chiroptera Cetartiodactyla Cetartiodactyla Carnivora Galliformes Carnivora

Rhabdoviridae Tobaniviridae Togaviridae

Rodentia Rodentia Perissodactyla Primates Diprotodontia Perissodactyla Chiroptera Diprotodontia Cetartiodactyla Cetartiodactyla Cetartiodactyla Carnivora Carnivora Passeriformes −2.5 0.0 2.5 5.0 −2.5 0.0 2.5 5.0 −2.5 0.0 2.5 5.0 log(Odds zoonotic)

Fig. S2. Partial effects plot for the family-specific reservoir effect in the top-ranked model in Fig. 2A. Points represent partial residuals, after accounting for the other effects in the model, while lines indicate the predicted effect. Shaded areas show the 95% confidence intervals of each predicted effect. Effects which are clearly separated from 0 (defined as having a confidence interval which does not cross 0) are highlighted in blue.

Nardus Mollentze and Daniel G. Streicker 3 of9 A Hantaviridae B Nairoviridae Rodentia Peribunyaviridae Primates Arenaviridae Perissodactyla Phenuiviridae Lagomorpha Orthomyxoviridae Diprotodontia Filoviridae Chiroptera Paramyxoviridae Cetartiodactyla

Pneumoviridae Reservoir Carnivora Rhabdoviridae Passeriformes Arteriviridae Galliformes Coronaviridae Anseriformes Tobaniviridae Reoviridae −2 −1 0 1 2 Astroviridae Caliciviridae log(Odds zoonotic) Flaviviridae C Hepeviridae Perissodactyla

Virus Picornaviridae Carnivora Togaviridae Cetartiodactyla Adenoviridae Chiroptera Asfarviridae Rodentia Herpesviridae Lagomorpha Papillomaviridae Primates Polyomaviridae Poxviridae Diprotodontia Hepadnaviridae Anseriformes Galliformes Reservoir phylogeny Passeriformes Retroviridae 0 200 400 600 −2 −1 0 1 2 Distance (My) log(Odds zoonotic)log(Odds zoonotic) 0.00 0.25 0.50 0.75 1.00 1.25−4 −2 0 2 4 log(Odds zoonotic)log(Odds zoonotic)

Fig. S3. Effect of virus taxonomy and reservoir phylogeny on the probability of being zoonotic, after accounting for all other effects in the model. (A) Partial effects for the individual families making up the taxonomic random effect, in the best model containing a virus taxonomy effect in the absence of any reservoir-related random effects (ranked 34th in Fig. 2; ∆AIC = 14.1). (B-C) Partial effects for the reservoir and reservoir phylogeny random effects from the top-ranked model in which each occurred without a virus taxonomy random effect (ranked 127th and 128th, respectively; ∆AIC = 30.8 in both cases). Bars indicate partial effects, after accounting for all other variables in the model, with shaded regions showing the extent of their 95% confidence intervals, while points indicate partial residuals for individual virus species. Families with effects clearly separated from 0 are highlighted in blue.

4 of9 Nardus Mollentze and Daniel G. Streicker A B E Deviance Rank ∆AIC explained Virus-related publications 1 0.00 56.4 2 1.81 57.1 140 3 1.92 56.7 Reservoir sp. richness 130

4 1.97 38.2 120 5 2.64 44.9 Phylogenetic distance 110 6 3.81 57.1 100 7 3.87 38.7 Mammalian reservoir 8 4.36 46.3 90 9 5.63 15.8 -1 0 1 2 80 10 5.71 0.0 Coefficient estimate 70 11 6.31 24.6 C D 60 12 6.50 9.4 6 13 7.15 4.5 5 50 14 7.61 28.9 40 Predicted number of virus species 15 7.62 15.8 4 30 16 8.46 9.8 4 20

10 of virus species Effect on number of virus3 species

2 Effect on number 0

2 3 4 5 6 7 8 9 0 10 20 30 40 50 60 70 80 90 100 110

Mammalian reservoir log(Reservoir spp. richness) Observed number of virus species Phylogenetic distance Phylogenetic Reservoir sp. richness Reservoir sp. Virus−related publications

Fig. S4. Factors predicting the total number virus species maintained by different animal orders (viral richness). (A) Models for all possible variable combinations ranked by AIC. Each row represents a model, while columns represent variables. Filled cells and white cells indicate variable inclusion and absence, respectively. The top four model are colour coded, with colours re-used in all other panels to identify the respective models. (B) Coefficient estimates for the top 4 models: points indicate the maximum likelihood estimate; lines show 95% confidence intervals. All variables were scaled by dividing them by 2 times their standard deviation. (C–D) Partial effect plots for variables in the top model. Lines and shading indicate the partial effects and 95% confidence intervals, with points showing partial residuals. (E) Predicted viral richness for each reservoir group when using the top model (blue in panel A).

Nardus Mollentze and Daniel G. Streicker 5 of9 A

60

Model rank

1 40 2

3

4 20 Predicted number of zoonotic species

0

0 10 20 30 40 0 10 20 30 40 0 10 20 30 40 0 10 20 30 40 Observed number of zoonotic species

B

140

120

100 Model rank

1 80 2

60 3 4 40 Predicted number of virus species 20

0

20 40 60 80 100 20 40 60 80 100 20 40 60 80 100 20 40 60 80 100 Observed number of virus species

Fig. S5. Predictions from the top four models predicting the number of zoonotic species and the total viral richness. (A) Observed and predicted number of zoonotic species for each reservoir group when using each of the top four models in Fig. 4A. (B) Observed and predicted total viral richness per reservoir group, for the top four models illustrated in Fig. S4A.

6 of9 Nardus Mollentze and Daniel G. Streicker hr umrsdt h ottxnmcodrlvl eann nqecmiain fvrsseiswt aho h eevi resicue ntepiayaayi.These analysis. primary the in included orders reservoir the of each with species virus of combinations unique retaining level, order taxonomic the to summarised where iu ae otems eetIT-prvdtxnm Mse pce it21b tp:/cvgoa/lsmse-pce-it/,PRcnre iu eetosfo ( from detections virus PCR-confirmed https://ictv.global/files/master-species-lists/), 2018b, List Species (Master taxonomy ICTV-approved recent most the to names virus adsMleteadDne .Streicker G. Daniel and Mollentze Nardus prediction. grey. this in of shaded interval are confidence variable 95% predictor the predicted the indicating the of range indicating regions with entire lines shaded (A), the with and F, in over value, – shown 0 observed B are cross mean panels AIC interval their in by confidence at ranked illustrated 95% variables models are whose other ( 30 model Effects all of top top keeping residuals. dataset The the when partial the in them. x-axis, illustrate Because present including Points the text. effects when on main The similar variable the variables. were the in representing results of described columns although effect analyses and data, the models reservoir repeating individual the before representing from interactions rows removed reservoir-virus were known these of hosts, list avian the contain to added were interactions host-virus S6. Fig. Rank A 13 12 11 10 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 9 8 7 6 5 4 3 2 1 Deviance explained (%) rdcoso ontcpoest hnicuighssi hc iue aebe eetd u o hc aneac a o encnre.Atrupdating After confirmed. been not has maintenance which for but detected, been have viruses which in hosts including when propensity zoonotic of Predictors iu effects Virus Virus publications Cytoplasmic replication Vector−borne Virus taxonomy Reservoir spp. richness eevi effects Reservoir Reservoir publications

Phylogenetic distance 0 4 Reservoir phylogeny 8 Reservoir group 12 Reservoir group

(family−specific) 16 ∆ 9.5048.8% 9.3048.9% 9.1840.1% 9.1640.2% 9.1640.2% 9.1440.1% 8.7548.3% 8.5248.5% 8.5040.3% 8.4640.3% 8.4640.3% 8.3650.6% 8.2348.6% 7.8340.1% 7.7840.1% 7.7740.1% 7.5648.2% 7.5050.3% 5.1149.4% 4.2149.1% 3.2949.9% 2.9750.6% 2.5950.3% 2.4149.5% 2.2849.6% 2.1949.7% 1.3749.3% 1.2449.4% 1.1349.4% 0.0049.2% AIC explained Deviance B C F Virus taxonomy log(Odds zoonotic) log(Odds zoonotic) −2.5 −1 0.00 0.0 2.5 5.0 7.5 0 1 2 log(Virus publicationlog(Virus count) 0 0.25 3 0.50 Vector− borne Retroviridae Parvoviridae Circoviridae Hepadnaviridae Poxviridae Polyomaviridae Papillomaviridae Herpesviridae Asfarviridae Adenoviridae Togaviridae Picornaviridae Hepeviridae Flaviviridae Caliciviridae Astroviridae Reoviridae Tobaniviridae Coronaviridae Arteriviridae Rhabdoviridae Pneumoviridae Paramyxoviridae Filoviridae Orthomyxoviridae Phenuiviridae Arenaviridae Peribunyaviridae Nairoviridae Hantaviridae 6 0.75 log(Odds zoonotic) 1.00 9 1.25 D E log(Odds zoonotic) log(Odds zoonotic) −2 −1 −2 0 1 2 3 0 2 −2 o(dszoonotic) log(Odds Phylogenetic distance(My) 0 Carnivora 0

Cetartiodactyla 100

Perissodactyla 2 200 Primates 4

1 Rodentia i not did ) 300 9 of 7 Retroviridae 1 ) 9 of 8 grey. illustrate in Points shaded prediction. variable are this variable the of predictor of interval effect the confidence predicted of 95% the range the indicating entire indicating lines the regions with over F, shaded 0 – and cross B value, interval panels observed confidence in mean 95% illustrated their whose are at Effects model variables residuals. top other the partial all in keeping present when effects x-axis, The the variables. on representing columns and models individual S7. Fig. A Rank 18 17 16 15 14 13 12 11 10 30 29 28 27 26 25 24 23 22 21 20 19 1 9 8 7 6 5 4 3 2 Deviance explained (%) rdcoso ontcpoest hnuigdt rmmmainascae iue ny h o 0mdl akdb I r hw n() ihrw representing rows with (A), in shown are AIC by ranked models 30 top The only. viruses mammalian-associated from data using when propensity zoonotic of Predictors iu effects Virus Virus publications Cytoplasmic replication Vector−borne Virus taxonomy Reservoir spp. richness eevi effects Reservoir Reservoir publications

Phylogenetic distance 0

Reservoir phylogeny 10 Reservoir group 20 Reservoir group (family−specific) ∆ 13.2555.2% 12.7354.8% 10.7553.2% 9.8853.3% 9.5653.5% 9.5454.3% 8.8353.7% 8.5054.2% 7.9053.2% 7.7453.6% 5.6153.1% 4.4354.6% 2.3152.5% 2.2853.0% 0.8153.7% 0.7953.2% 0.3553.1% 0.0053.2% 16.7141.3% 16.6642.3% 16.6542.3% 16.6542.3% 16.1655.1% 16.1154.9% 15.7854.8% 15.6240.1% 15.6240.1% 15.6240.1% 15.4853.2% 13.6955.2% AIC explained Deviance D B C E log(Odds zoonotic) Virus taxonomy log(Odds zoonotic) log(Odds zoonotic) 0.0 −2 −1 −1 0 4 8 0 1 2 3 0 1 2 3 4 5 log(Virus publicationlog(Virus count) 0.1 Phylogenetic distance(My) 0 Cytoplasmic 0 0.2 replication log(Odds zoonotic) 0.3 o(dszoonotic) log(Odds 0.4 3 100 0.5 −2 6 200 −1 Vector− borne 0 adsMleteadDne .Streicker G. Daniel and Mollentze Nardus 9 300 1 2 F log(Odds zoonotic) −2 −2 −2 −2 4 0 2 4 0 2 4 0 2 4 0 2

Carnivora Cetartiodactyla Chiroptera Diprotodontia Perissodactyla Primates

Rodentia

Adenoviridae Rhabdoviridae Retroviridae Picornaviridae References 1. KJ Olival, et al., Host and viral traits predict zoonotic spillover from . Nature 546, 646–650 (2017).

Nardus Mollentze and Daniel G. Streicker 9 of9