*2.3. Population Genetic Diversity*

We calculated the probability that each collection was in Hardy–Weinberg equilibrium (*HW-P*) using the chi-square test in the R package Genepop v1.0.5 (Montpellier, France) [25] with probabilities >0.05; for collections in Hardy–Weinberg equilibrium, expected heterozygosity (*He*) and observed heterozygosity (*Ho*) were then calculated. We calculated nucleotide diversity (*Pi*) for each of the five collections using VCFtools v0.1.14 (Cambridge, UK) [26]. We also used Genepop v1.0.5 (Montpellier, France) [25] to calculate the inbreeding coefficient within each collection (*Fis*), the inbreeding coefficient across all collections (*Fit*), and the pairwise genetic differentiation coefficient between collections (*Fst*). We calculated Reynolds's genetic distance (*DR*) and gene flow (*Nm*) between pairs of collections as −ln (1 − *Fst*) and (1 − *Fst*)/4 × *Fst*, respectively.

### *2.4. Population Genetic Structure*

We used PLINK 1.9 (Massachusetts, USA) [27] to assess SNPs across the entire constructed reference sequence of the yellowtail kingfish genome and to identify SNPs with no close linkages (using the main parameter "indep-pairwise 50 10 0.6", where 0.6 is the r2 threshold). We used ADMIXTURE v1.3.0 (Houston, USA) [28], with 10 different seeds for 10 repeated analyses, to determine the most likely number of subpopulations (K) among the 143 individuals, with tested values of K from 1 to 10. Using PLINK v2.0 [29], we performed a PCA (principal component analysis) of the obtained SNP markers. Finally, we constructed a phylogenetic tree based on neighbor-joining (NJ) analyses using bottom-up clustering for 143 individual yellowtail kingfish, using TreeBest v1.9.2 (Hinxton, UK) [30] with 1000 bootstrap replicates.

#### **3. Results**

#### *3.1. Yellowtail Kingfish Population Genetics*

The average number of unique tags per sample was 287,594, and the average sequencing depth was 27.13×. Across all samples, the percentage of unique tags was 94.61–96.25%. After quality control filtering, 48,710 sites remained.

In the genetic diversity indicators, *He* ranges from 0 to 1, where 0 represents no polymorphism, and 1 represents an infinite number of alleles with the same frequency. A higher *Pi* value indicates greater genetic diversity between sequences, meaning there is a larger degree of variation among the samples or populations. Conversely, a smaller *Pi* value indicates higher genetic similarity between sequences, implying a lower level of variation among the samples or populations. *Fis* represents the inbreeding coefficient within populations. *Fit* represents the total population inbreeding coefficient, and their values range from −1 to 1. A significantly positive value of *Fis* indicates a high level of inbreeding within the population, while a significantly negative value suggests the presence of outbreeding. *Fst* ranges from 0 to 1, and a larger value indicates a more pronounced genetic differentiation between subpopulations. Across all five yellowtail kingfish collections, the average probability of the Hardy–Weinberg equilibrium was 0.4112 (*p* > 0.05; Table 1), which indicated that each of the collections were in Hardy–Weinberg equilibrium. The average *Ho* was 0.0824, average *He* was 0.2013, average *Pi* was 0.2020, average *Fis* was 0.0627, average *Fit* was 0.3298, and average *Fst* was 0.2898.

**Table 1.** Genetic diversity of five yellowtail kingfish collections.


Notes: A Japanese wild collection (JW), Chinese farmed collection (CF), Chinese wild collection (CW), Australian farmed collection (AF), and Australian wild collection (AW). *Ho*: observed heterozygosity; *He*: expected heterozygosity; *Pi*: nucleotide diversity; HW-P: *p*-value for the Hardy–Weinberg equilibrium test.

In general, the *Ho*, *He*, and *Pi* values for the farmed Chinese yellowtail kingfish collection (0.0828, 0.0814, and 0.0829, respectively) were greater than those for the wild Chinese yellowtail kingfish collection (0.0718, 0.0771 and 0.0785, respectively, *p* < 0.05). In contrast, the *He* and *Pi* values for the farmed Australian yellowtail kingfish collection (0.0927 and 0.0943, respectively) were lower than those for the wild Australian yellowtail kingfish population (0.0955 and 0.0974, respectively), although the value of Ho was greater for the farmed collection (0.0940) for the wild (0.0902; Table 1).

Among the Chinese and Japanese collections, *Fst* was 0.00097–0.01888 and *DR* was 0.0010–0.0191; similarly, *Fst* and *DR* were 0.02569 and 0.0260, respectively, between the two Australian collections (Table 2), suggesting no genetic differentiation (*Fst* < 0.05). However, the *Fst* values between the Chinese or Japanese collections and the Australian collections were 0.7256–0.7447 (*DR*: 1.2932–1.3653; Table 2), suggesting substantial genetic differentiation (*Fst* > 0.25).

**Table 2.** Pairwise measures of genetic differentiation among five collections of yellowtail kingfish.


Notes: A Japanese wild collection (JW), Chinese farmed collection (CF), Chinese wild collection (CW), Australian farmed collection (AF), and Australian wild collection (AW). The genetic differentiation coefficient (*Fst*) is shown below the diagonal, and the genetic distance (*DR*) is shown above the diagonal.

The *Nm* values among the Chinese and Japanese collections were high (12.9287–256.4499), as was the *Nm* value between the Australian collections (9.9002; Table 3). This suggested frequent gene exchange between the wild Chinese and farmed Chinese and wild Japanese collections, as well as between the wild Australian and farmed Australian collections. Indeed, the highest *Nm* was calculated between the Chinese and Japanese wild collections, and the gene flow may be frequent and/or recent. In contrast, the *Nm* values between the Chinese or Japanese and Australian collections were low (0.0864–0.0953; Table 3), indicating little to no gene exchange between the Asian and Australian yellowtail kingfish collections.

**Table 3.** Values of the gene flow parameter (*Nm*) among the five collections of yellowtail kingfish.


Notes: A Japanese wild collection (JW), Chinese farmed collection (CF), Chinese wild collection (CW), Australian farmed collection (AF), and Australian wild collection (AW).

#### *3.2. Yellowtail Kingfish Population Structure*

The ADMIXTURE cross-validation error levels indicated that the most likely number of subpopulations (parameter *K*) was 2 (Figure S1). When *K* = 2, the Australian wild and farmed collections are grouped together in one cluster, while the Chinese wild and farmed collections, as well as the Japanese wild collections, form another cluster. When *K* = 3 or *K* = 4 (Figure 2B,C), individual outliers from the cultured population cluster separately within the Australian and Chinese farmed collections, but did not show distinct separations. Based on the current analysis, the origins of the Chinese and Japanese populations are similar, while the Australian collections origins are different (Figure 2A). Similarly, our PCA showed that the Chinese and Japanese collections clustered together, distinct from the Australian cluster (Figure 3A); PC1 and PC2 explained 41.16% and 3.73% of the variance across populations, respectively. On PC1, there were individual outliers within the wild Australian collection. PC3 revealed a weak clustering pattern in the Australian cultured collection, and PC3 explained only 1.26% of the variance across populations (Figure 3B). Similarly, PC4 indicated a weak clustering pattern in the Chinese cultured collection, and PC4 explained only 1.16% of the variance across populations (Figure 3C).

The phylogenetic tree shows that Australian collections and Chinese and Japanese collections are divided into two distinct clades. The farmed and wild Australian collections formed two distinct clades. Whereas, the wild Chinese and Japanese collection formed a branch, with the wild Japanese collection branching off from the farmed Chinese collection, though no clear clade divergence was observed among these three collections (Figure 4).

**Figure 2.** (**A**). The genetic structure of the five yellowtail kingfish collections for *K=2*. (**B**). The genetic structure of the five yellowtail kingfish collections for *K = 3.* (**C**). The genetic structure of the five yellowtail kingfish collections for *K* = 4. Notes: Each bar represents one individual. Different colors suggest different origins and show the proportion of each genotype belonging to each genetic cluster.

**Figure 3.** *Cont.*

**Figure 3.** (**A***–***C**). Principal components analysis (PCA) showing the genetic structure of five yellowtail kingfish collections. Notes: A Japanese wild collection (JW), Chinese farmed collection (CF), Chinese wild collection (CW), Australian farmed collection (AF), and Australian wild collection (AW). Each symbol represents an individual fish.

**Figure 4.** Neighbor-joining phylogenetic tree of the five yellowtail kingfish collections. Notes: The reliability of the phylogenetic tree was assessed using the bootstrap method with 1000 replicates. Branch lengths reflect the genetic distance. A Japanese wild collection (JW), Chinese farmed collection (CF), Chinese wild collection (CW), Australian farmed collection (AF), and Australian wild collection (AW).
