Fig. 4: Accuracy of the multi-generation implicit dispersal model in explaining compiled genetic structures.

a MLPE linear mixed model R2 between \({F}_{{{{{{\rm{st}}}}}}}^{{{{{{\rm{mod}}}}}}}\) and \({F}_{{{{{{\rm{st}}}}}}}^{{{{{{\rm{obs}}}}}}}/(1-{F}_{{{{{{\rm{st}}}}}}}^{{{{{{\rm{obs}}}}}}})\). Filled dots highlight the 37 significant correlations (p value*). Note that some results reported by a given study are analyzed separately: (i) Weber et al.87 used nuclear DNA marker (1) and mtDNA marker (2); (ii) Carreras et al.58 considered all the loci together (3) and then only the Mediterranean outliers loci (4); (iii) Marzouk et al.88 analyzed nuclear DNA marker (5) and mtDNA marker (6). Error-bars for all studies represent the 95 % confidence intervals. b Network representation of \({F}_{{{{{{\rm{st}}}}}}}^{{{{{{\rm{obs}}}}}}}/(1-{F}_{{{{{{\rm{st}}}}}}}^{{{{{{\rm{obs}}}}}}})\) (observed \({F}_{{{{{{\rm{st}}}}}}}\), left) and \({F}_{{{{{{\rm{st}}}}}}}^{{{{{{\rm{mod}}}}}}}\) (modelled \({F}_{{{{{{\rm{st}}}}}}}\), right) and their corresponding scatterplot for the flathead grey mullet (Mugil cephalus;61 red dot in a). c same as b but for a seagrass (Cymodocea nodosa;62 red dot in a). Source data are provided as a Source Data file. Icons credit: © vectors market, © Agne Alesiute, © Elisabetta Calabritto, © Luis Prado, © Joi Stack, © Tatina Vazest, © mindgraphy, © Oleksandr Panasovskyi, © ProSymbols, © Sean Maldjian (changes were made on all the icons, Creative Commons BY 4.0 license).