QST–FST comparisons: evolutionary and ecological insights from genomic heterogeneity

“…prediction of the expectation for quantitative traits under neutrality which, thus, solidified the inference that quantitative traits typically show greater genetic divergence among populations than expected under neutrality (Meril¨a & Crnokrak, 2001; De Kort et al., 2013; Leinonen et al., 2013).”

“The neutrality issue can partially be addressed by using multiple markers randomly sampled across the whole genome (Leinonen et al., 2013).”

“This is in contrast to the results of most quantitative genetic studies, which reported divergent selection (De Kort et al., 2013; Leinonen et al., 2008, 2013; Merilä & Crnokrak, 2001).”

“Consequently, differences between the two indices can be attributed to selection (Leinonen et al., 2013).”

“QST values greater than FST are interpreted as evidence for divergent natural selection, and when values are similar, selection and genetic drift cannot be differentiated (Leinonen et al. 2013a; Whitlock 2008).”

“This second approach, follows the same premise as QST-FST comparisons (Leinonen et al. 2013), and asks whether there is more quantitative phenotypic divergence among-populations than expected from levels of (putatively) neutral molecular divergence.”

“Finally, introgression can also bias others tests requiring a neutral baseline of differentiation such as quantitative trait analyses (e.g. Leinonen et al. 2013).”

“A Q st-F st comparison framework was used to identify metabolites subject to divergent selection ( 95 ).”

“To do so, we compared neutral genetic differentiation ( 𝐹 𝑠𝑡 ) to its phenotypic analogues: 𝑄 𝑠𝑡 and 𝑃 𝑠𝑡 , which assess differentiation in quantitative traits (i.e., RMR, morphometric traits, growth, SGR) [66,67].”

“The morphological differentiation was similar to the molecular expectation [24].”

“By contrast, the genetic structure of neutral genes mainly reflects demographic processes such as genetic drift and changes in effective population size, which are the result of historical biotic and abiotic conditions throughout a species’ evolutionary history (Leinonen et al ., 2013).”

“PST is a measure of phenotypic differentiation between groups, comparable to F ST (Leinonen et al. 2013; Pujol et al. 2008).”

“If Q ST > F ST , it suggests adaptive divergence, while Q ST < F ST may imply spatially-homogeneous global adaptation (c.f. [3]; but note that [9,11] use ‘balancing selection’ when referring to this scenario).”

“Our observation that phenotypic differentiation exceeds neutral genetic differentiation in the presence of polymorphism and resource differentiation (EST > 0) aligns with the common notion that signals local adaptation (e.g. Whitlock, 2008; Leinonen et al., 2013).”

“…respectively; h2 expresses the heritability (the proportion of phenotypic variance that. is due to additive genetic effects); the scalar c expresses the proportion of the total variance that is presumed to be due to additive genetic variance across populations (Broker,2011; Leinonen et al., 2013).”

“All rights reserved To avoid misleading interpretation of QST-FST comparisons, confidence intervals of both indexes calculated on bootstrapped data have been taken into account (Leinonen et al., 2013).”

“Adapted from LEINONEN et al. (2013).”

“However, comparisons to neutral-marker F st must be interpreted with caution, because P st also depends on environmental variance and the ratio of additive genetic variances within versus among groups [23,24].”

“Unveiling the forces that maintain and generate pheno - typic variation is a main goal for evolutionary biologists (Leinonen et al. 2006 , 2013 ; Mazer and Damuth 2001 ).”

“If Q ST > F ST , it suggests adaptive divergence, while Q ST F ST may imply balancing selection [Spitze, 1993, Leinonen et al., 2013].”

“Phenotypic traits with Qst signi�cantly larger than the Fst estimated from neutral markers are considered as being under local adaptation, whereas Qst = Fst is the expectation under neutrality (Leinonen et al., 2013).”

“…phenotypic differentiation is caused by selective versus neutral processes, several studies – as far as we know, none strictly on aquatic plants – have compared phenotypic divergence (Pst) and differentiation of neutral loci (Fst), i.e. Pst – Fst comparisons (Brommer, 2011; Leinonen et al., 2013).”

“We tested whether phenotypic divergence (Pst) between these sister species deviated from expectations for quantitative traits under neutrality (Leinonen et al. 2013).”

“Similar studies have shown that morphological and agronomical traits divergence have been occurred in many taxa (See for example, Leinonen et al. 2013; Stojanova et al. 2018; Caré et al. 2018).”

“Coupling our QST approach with a neutral molecular marker study (FST) would help distinguish between these evolutionary processes (Leinonen et al., 2013) and thus provide a better understanding of the lack of genetic differentiation in this species.”

“It further allows quantifying the contribution of adaptive differentiation and random demographic processes to observed genetic differentiation patterns (Leinonen et al. 2013).”

“Some studies have raised concerns regarding biases in Qst and Fst estimation and in their comparison (Leinonen et al. 2013).”

“…values are therefore interpreted as evidence for directional, locally adaptive differentiation, whereas Q ST values within the neutral distribution are consistent with uniform or stabilizing selection across populations (Lande, 1992; Spitze, 1993; McKay and Latta, 2002; Leinonen et al., 2013).”

“…occur when P ST > F ST , favouring “extreme” phenotypes, while stabilizing selection occurs when Pst < Fst, favouring similar phenotypes across populations (Brommer 2011; Leinonen et al. 2008, 2013; Marin et al. 2020; Martin et al. 2008; Merilä & Crnokrak 2001; Seymour et al. 2019; Whitlock 2008).”

“This comparison enabled us to distinguish between natural selection and genetic drift as causes of population differentiation in a phenotypic trait (Leinonen et al., 2013).”

“Briefly, this comparison provides a means to disentangle the causes of population divergence, with 𝑄 𝑠𝑡 values higher, lower, or similar to 𝐹 𝑠𝑡 indicating divergent selection, stabilizing selection, or genetic drift, respectively (Leinonen et al. , 2013).”

“…phenotypic divergence often exhibits only partial or non-linear correspondence with genome-wide differentiation because it is additionally shaped by ecological context, developmental plasticity, and sex-specific processes (Leinonen et al., 2013; Pfennig et al., 2010; West-Eberhard, 2003).”

“…for example by determining whether the magnitude of additive genetic variation in a quantitative traits among population is different from neutral expectations (e.g. comparing Q st and F st ) (Antoniazza et al. 2010; Leinonen et al. 2013; Savolainen et al. 2007; Whitlock & Guillaume 2009).”

“…to measure additive genetic variance and thus formally compare phenotypic trait differentiation due to selection from that due to genetic drift (Leinonen et al. 2013), the differentiation measured in Scotti-Saintagne et al. (2019) on a slightly geographically wider set of provenances ( F ST =…”

“Individual Q ST values were calculated using the following model, where (cid:0)(cid:2) represents between-population variance and (cid:3)(cid:2) represents within-387 population (i.e., genotype-level) variance 24,25 : stabilizes around a particular phenotype 24-26 .”

“…with spatially heterogeneous selection in the Q_ST–F_ST framework when phenotypic differentiation exceeds neu - tral expectations, while recognizing that explicit genomic benchmarks and common-garden designs are required for formal inference (Chung et al. 2023; Leinonen et al. 2013; Whitlock 2008).”

“…to quantify the strength of spatially divergent selection, providing a stan-131 dardized measure of selection efficacy across simulation scenarios (Leinonen et al., 2013) and the level of local adaptation was estimated by the correlation between individual phenotypic values ( g ) and their local…”

“…been explored by comparisons of phenotypic and genomic differentiation between populations, usually expressed by the genetic variance among populations relative to the total variance of all populations sampled, i.e. the fixation index, F ST , or its phenotypic analogue, Q ST (Leinonen et al. 2013).”

“F ST comparison has been carried out on previous studies (Whitlock and Guillaume 2009; Leinonen et al. 2013; Colautti and Lau 2015; Boaventura-Novaes et al. 2018).”

“…garden studies combined with neutrality tests (e.g., Lande 1976; Rogers 1986; Spitze 1993; Merilä & Crnokrak 2001) provide a fairly straightforward way to differentiate adaptive vs. non-adaptive causes of phenotypic differentiation (Whitlock 2008; Leinonen et al. 2013; Savolainen et al. 2013).”

“Without selection, divergence in quantitative traits should equal the divergence at neutral marker loci (Leinonen et al., 2013, 2008).”

“There is a tendency for dominance variance to bias Q ST estimates, potentially leading to incorrect inferences of neutrality 39 .”

“While such trait-climate associations can indeed suggest patterns of local adaptation due to spatially varying selection (Leinonen et al. 2013), their interpretation can still be doubtful, because the climatic clines at which provenances occur often show parallelism with re-colonization routes…”

“FST values were estimated with 16 microsatellite markers assessed by Alberto et al. (2010), for which at least 20 trees in each population in situ were sampled in 2004 and 2005.”

“Ces expériences ont été couplées à une comparaison PST - FST qui a mis en évidence que les variations phénotypiques observées sur l’expérience conduite à Marseille ne pouvaient pas être le résultat de processus neutres uniquement, contrairement à l’expérience conduite en Corse (dans certains cas d’organi…”

“One commonly used approach for investigating the occurrence of selection is the FST vs. QST comparison (Leinonen et al. 2013).”

“Such a pattern is indeed consistent with local adaptation, assuming that (among other considerations such as the recency of selection) mutation rates are considerably lower than migration rates in these populations (Whitlock 1999; Hendry 2002; Leinonen et al. 2013).”

“QST behaves like a single-locus FST estimate (Leinonen et al. 2013); therefore, I compared PST across the distribution of single-locus FST estimates (see Population Structure) from the ‘neutral’ set of loci obtained after removing loci that showed high statistical correlations with body mass and…”

“Interestingly, such data are available (Leinonen et al. 2013); however, they have yet to be synthesized and incorporated into models of food-web dynamics.”