Abstract:Heritable genetic versus non-heritable plastic phenotypic variation determine how selection may shape evolutionary responses. Population genomic studies typically aim to identify associations between loci (SNPs) and phenotypes, missing broader insight into heritable basis of polygenic traits with quantitative genomic approaches. Genomic animal models can facilitate estimates of trait heritability from genomic data but can be challenging to apply to wild organisms. This is because the populations of many wild organisms are large, comprising many distantly related individuals, which makes the sampling close-kin unlikely. This in turn limits the variation of relatedness in a sample, reducing the efficacy of genomic animal models. We developed a novel framework for applying genomic animal models to wild organisms where sampling close-kin may be near impossible. Our framework combines resampling of genetic markers with simulating different levels of heritability to infer the power and reliability of genomic animal models fitted to datasets lacking kin structure. We use our framework to estimate heritability of head shape traits in a wild marine intertidal fish, Bathygobius cocosensis (Bleeker 1854), using ddRAD SNP genotypes and phenotypes measured using geometric morphometric methods. In its eastern Australian range, B. cocosensis is highly abundant, has high effective gene flow, and occupies diverse habitats and microhabitats. Placing our observed genomic animal model results in the context of simulated results, we conclude that two of the head shape traits were heritable (estimated h2 = 0.26 and 0.18). This suggests that head shape phenotypes could evolve through heritable genetic changes in B. cocosensis, although there is likely a large non-heritable component. We also found strong phenotypic differentiation (pairwise PST ≤ 0.33) despite very weak genetic differentiation (pairwise FST < 0.002), suggesting that environmental factors might structure phenotypes amidst processes homogenising genetic variation. Our study addresses key methodological gaps for using quantitative genomic approaches on wild organisms, offering a rare example of partitioned heritable and non-heritable contributions to phenotypic variation in a non-model wild marine fish. This would have been difficult without our new framework to overcome the challenges of a dataset lacking kin structure.

What problem does this paper attempt to address?

### Problems the paper attempts to solve This paper aims to solve the methodological challenges encountered when applying genomic animal models in non - model organisms in the wild, especially for those data sets lacking a close - relative structure. Specifically, the main problems studied can be summarized as follows: 1. **Methodological problems**: - How to use genomic animal models to partition phenotypic variation in non - model organisms, especially in the absence of a detectable close - relative structure? - By resampling genetic markers and simulating different levels of heritability, how to evaluate the effectiveness and reliability of genomic animal models in wild organisms? 2. **Biological problems**: - How are genetic variation and phenotypic variation distributed in *Bathygobius cocosensis* (a marine fish living in the intertidal zone) at different spatio - temporal scales? - What are the relative contributions of genetic variation and non - genetic variation to the head - shape phenotype of *Bathygobius cocosensis*? ### Specific methods To answer the above questions, the researchers developed a new framework, combining genetic marker resampling and simulation techniques. The specific steps are as follows: 1. **Observation data**: - Randomly draw multiple SNP sets (J sets) from the total genetic marker pool after quality filtering and redundancy reduction. - For each trait (i) and each SNP set (j), estimate the corresponding genomic relationship matrix A, and use this matrix to fit a genomic animal model, including any covariates as fixed effects. - Fit a control model without A. - By iterating through J SNP sets, obtain a series of genomic animal model estimates that are applicable to the same phenotype and fixed effects but have different As. - Calculate the heritability estimate (h²) for each trait and the range of P - values for model comparison tests to evaluate the reliability and power of estimates based on genetic marker selection. 2. **Simulation data**: - On the basis of each trait and each SNP set, use the `family_sim_qtl` function in the R package `genoMalicious` to simulate different levels of heritability. - Through simulated data, obtain the expected estimates and their variability at different heritability levels, thereby providing a context for the observed results to understand the expected estimates and their variability under a given heritability level and experimental design. ### Conclusions Through this method, the researchers were able to evaluate the effectiveness and reliability of genomic animal models in wild organisms lacking a close - relative structure and further explore the genetic basis of the head - shape phenotype in *Bathygobius cocosensis*. The results showed that two head - shape traits are heritable (with estimated heritabilities of 0.26 and 0.18 respectively), indicating that these phenotypes can evolve in *Bathygobius cocosensis* through genetic changes. However, there is also a large non - genetic component. In addition, although the genetic differentiation is weak (FST < 0.002), the phenotypic differentiation is strong (PST ≤ 0.33), suggesting that environmental factors may play an important role in the phenotypic structure.

A novel framework for assessing the robustness of genomic animal models applied to a wild marine fish

Dissecting the genetic architecture of quantitative traits using genome‐wide identity‐by‐descent sharing

Genomic islands of divergence and their consequences for the resolution of spatial structure in an exploited marine fish

Cross-species Exon Capture and Whole Exome Sequencing: Application, Utility and Challenges for Genomic Resource Development in Non-model Species

Investigating pedigree- and SNP-associated components of heritability in a wild population of Soay sheep

Genetic adaptation despite high gene flow in a range‐expanding population

Accuracy of within-family multi-trait genomic selection models in a sib-based aquaculture breeding scheme

Genomic selection models double the accuracy of predicted breeding values for bacterial cold water disease resistance compared to a traditional pedigree-based model in rainbow trout aquaculture

Signatures of polygenic adaptation associated with climate across the range of a threatened fish species with high genetic connectivity

Simulation Analysis on Genomic Selection of Grouper (epinephelus Coioides) Breeding for Categorical Traits

Species-specific Genetic Variation in Response to Deep-Sea Environmental Variation Amongst Vulnerable Marine Ecosystem Indicator Taxa.

The between-population genetic architecture of growth, maturation, and plasticity in Atlantic salmon

Ontogenetic variation in the marine foraging of Atlantic salmon functionally links genomic diversity with a major life history polymorphism

High Resolution Genomes of Multiple Xiphophorus Species Provide New Insights into Microevolution, Hybrid Incompatibility, and Epistasis.

Genomic evidence for local adaptation in the ovoviviparous marine fish Sebastiscus marmoratus with a background of population homogeneity

Three chromosomal rearrangements promote genomic divergence between migratory and stationary ecotypes of Atlantic cod

Microbiome and epigenetic variation in wild fish with low genetic diversity

Estimating Realized Heritability for Growth in Zhikong Scallop (chlamys Farreri) Using Genome-Wide Complex Trait Analysis

Efficiency of genomic selection using Bayesian multi-marker models for traits selected to reflect a wide range of heritabilities and frequencies of detected quantitative traits loci in mice

Accounting for Population Structure and Phenotypes From Relatives in Association Mapping for Farm Animals: A Simulation Study

Towards population genomics in non-model species with large genomes: a case study of the marine zooplankton Calanus finmarchicus