Table 1.
Coefficients | 95% confidence intervals | |||
---|---|---|---|---|
Pest group (n) | β0 | β1 | β0 | β1 |
Bacteria (30) | 3.2613 | −2.9706 | 1.5581–4.3791 | −3.4605 to −2.2215 |
Fungi (95) | 4.3961 | −3.3249 | 3.1958–5.3539 | −3.7439 to −2.8019 |
Oomycetes (32) | 2.0763 | −2.6679 | −0.8020–3.6867 | −3.3745 to −1.4216 |
Insects (637) | 3.2441 | −2.7004 | 2.7108–3.7216 | −2.9073 to −2.4694 |
Mites (87) | 1.9584 | −2.1681 | 0.3266–2.9620 | −2.6039 to −1.4625 |
Mollusks (37) | −0.4667 | −1.1391 | −4.5587–1.5261 | −1.9896 to –0.5920 |
Nematodes (70) | 2.7157 | −2.6249 | 1.8376–3.4603 | −2.9498 to −2.2445 |
Viruses (108) | 8.4044 | −5.2014 | 7.7856–8.9129 | −5.4320 to −4.9236 |
Plants (53) | 1.9979 | −2.2775 | 1.0617–2.5609 | −2.5233 to −1.8644 |
Shown are the coefficients of logistic regressions (median and 95% confidence intervals) of whether target host genus was known to be susceptible (S) to a pest from a source host genus, as a function of the phylogenetic distance between source and target hosts. The regression takes the form of logit(S) = β0 + β1*log10(PD + 1), where PD is the phylogenetic distance (time of independent evolution in My) between the source and target host genera. All pest groups except mollusks had phylogenetic signals significantly different from zero (95%CI for β1 did not overlap 0). The probability that a target host is susceptible to a pest from a source host is then prob(susceptible) = exp[logit(S)]/[1 + exp(logit(S))]. Number of pest species included in the regression (those with >1 known host genus) is given in parentheses as n.