Fig 1.
Flow diagram with the steps in the analysis.
Fig 2.
Distribution of species, administration routes and duration across studies.
Fig 3.
Summary of largest effect sizes obtained (OR ≥ 2).
The plot contains all the target-tissue associations (regardless of mode of action) for which a meta-analysis produced an OR ≥ 2. Colours are based on p-values (yellow: not significant at p ≤ 0.05, orange: significant at p ≤ 0.05 but with FDR > 5%, red: significant) and with FDR ≤ 5%.
Fig 4.
Target profile forest plot for androgen receptor agonists (random effects).
The meta-analyses can be grouped by target to produce profiles of events per tissue. The figure shows the profile obtained for the meta-analyses involving androgen receptor agonists for each tissue analysed. The results of the meta-analyses allow to identify the tissues where there was a significant increase in number of animals with events vs controls.
Fig 5.
Androgen agonists subgroup analysis for the effect on number of events in testes (random effects): Forest plots with results by species and tests for subgroup differences.
Forest plots can be used to evaluate the consistency of the effect across covariates and studies.