DISCUSSION

The present study provides original insights by demonstrating, for the first time in Thai native chickens, that egg-laying persistency constitutes a partially independent biological and genetic dimension, separable from growth, sexual maturity, and total egg output, as revealed by PCA and bivariate animal models. This finding has important implications for sustainable breeding because most existing genetic evaluations in Pradu Hang Dum and similar breeds have prioritized cumulative egg number or heat stress resilience using reaction norms or random regression, often overlooking long-term laying stability.

Growth performance and egg production traits

The growth pattern of Thai native chickens observed in this study reflects a slow, steady growth rate, a defining characteristic of indigenous chickens raised under tropical conditions [4, 5]. This slow-growing phenotype represents an adaptive strategy that prioritizes heat tolerance and metabolic efficiency rather than rapid tissue accretion [34, 35]. Reproductive traits further reinforce the indigenous characteristics of this population, as the observed AFE (approximately 163 days) and AFE_WT (approximately 2032 g) are comparable to previous reports in Thai native chickens, which typically reach sexual maturity later than commercial layers because of slower physiological and endocrine development [2, 36]. The gradual increase in egg weight with advancing laying age reflects normal maturation of the reproductive system in indigenous chickens [7, 17]. A decline in egg-laying persistency during the late production phase is also characteristic of indigenous chickens and low-input production systems [12]. The high coefficient of variation for PR and the strong genetic correlations with egg number traits should be interpreted with caution because ratio-based persistency traits may be affected by part–whole dependency and denominator effects [37].

Traits | Genotypes of DRD2 gene | p-value | Genotypes of VIP gene | p-value
CC | CT | TT | DD | ID | II
BW0 | 0.08 ± 0.16 | 0.17 ± 0.15 | −0.09 ± 0.31 | 0.74 | 0.09 ± 0.21 | 0.10 ± 0.21 | 0.11 ± 0.15 | 0.99
BW4 | 0.20 ± 0.54 | 0.82 ± 0.51 | −0.31 ± 1.02 | 0.52 | 0.51 ± 0.69 | 0.23 ± 0.71 | 0.48 ± 0.49 | 0.95
BW8 | 0.72 ± 2.03 | 3.11 ± 1.92 | −0.98 ± 3.86 | 0.53 | 2.02 ± 2.60 | 0.68 ± 2.69 | 1.91 ± 1.85 | 0.92
BW12 | 1.52 ± 3.46 | 4.63 ± 3.27 | −1.00 ± 6.58 | 0.68 | 3.15 ± 4.43 | 0.99 ± 4.57 | 3.27 ± 3.14 | 0.91
BW16 | 1.71 ± 3.98 | 5.72 ± 3.76 | −1.91 ± 7.56 | 0.59 | 3.64 ± 5.09 | 1.65 ± 5.26 | 3.65 ± 3.62 | 0.95
AFE | −0.35 ± 0.22 | 0.05 ± 0.21 | 0.13 ± 0.41 | 0.34 | −0.47 ± 0.28 | 0.04 ± 0.29 | 0.01 ± 0.20 | 0.32
AFE_EW | 0.00 ± 0.03 | −0.01 ± 0.03 | 0.07 ± 0.05 | 0.37 | 0.02 ± 0.04 | −0.04 ± 0.04 | 0.02 ± 0.03 | 0.33
AFE_WT | 1.03 ± 5.10 | −1.41 ± 4.81 | 4.66 ± 5.67 | 0.33 | 4.72 ± 6.50 | −5.54 ± 6.72 | 5.02 ± 4.62 | 0.17
EN270 | 0.15 ± 1.21ᵃ | 1.76 ± 1.14ᵃ | −5.09 ± 2.29ᵇ | 0.03 | 4.22 ± 1.54ᵃ | −0.92 ± 1.59ᵇ | −1.10 ± 1.09ᵇ | 0.01
EN360 | 1.61 ± 0.35ᵃ | 0.52 ± 0.33ᵇ | −4.14 ± 0.67ᶜ | <0.0001 | 2.84 ± 0.46ᵃ | 0.17 ± 0.47ᵇ | −0.64 ± 0.32ᵇ | <0.0001
EW270 | −0.03 ± 0.06 | 0.05 ± 0.06 | 0.08 ± 0.12 | 0.58 | 0.06 ± 0.08 | −0.09 ± 0.09 | 0.05 ± 0.06 | 0.34
EW360 | −0.02 ± 0.05 | 0.04 ± 0.04 | 0.06 ± 0.09 | 0.57 | 0.05 ± 0.06 | −0.07 ± 0.06 | 0.03 ± 0.04 | 0.34
PLS | 0.19 ± 0.13 | −0.16 ± 0.12 | −0.29 ± 0.25 | 0.09 | −0.06 ± 0.17 | 0.01 ± 0.17 | −0.04 ± 0.12 | 0.96
PR | 0.88 ± 0.61 | −0.89 ± 0.58 | −1.14 ± 1.17 | 0.08 | −0.52 ± 0.79 | 0.15 ± 0.82 | −0.16 ± 0.56 | 0.83

Traits | Genotypes of NPY gene | p-value | Genotypes of MTNR1C gene | p-value
LL | MM | SS | AA | AG | GG
BW0 | 0.30 ± 0.32 | 0.26 ± 0.17 | −0.05 ± 0.14 | 0.32 | 0.13 ± 0.21 | 0.11 ± 0.15 | 0.06 ± 0.20 | 0.97
BW4 | 1.33 ± 1.06 | 1.16 ± 0.58 | −0.24 ± 0.48 | 0.12 | 0.45 ± 0.70 | 0.50 ± 0.50 | 0.28 ± 0.68 | 0.96
BW8 | 5.32 ± 3.99 | 4.59 ± 2.18 | −1.10 ± 1.79 | 0.08 | 1.62 ± 2.63 | 2.04 ± 1.88 | 0.93 ± 2.55 | 0.94
BW12 | 5.32 ± 3.99 | 4.59 ± 2.18 | −1.09 ± 1.79 | 0.06 | 2.48 ± 4.48 | 3.72 ± 3.21 | 0.99 ± 4.34 | 0.88
BW16 | 9.45 ± 6.78 | 8.00 ± 3.70 | −2.29 ± 3.05 | 0.09 | 3.23 ± 5.15 | 3.95 ± 3.69 | 1.67 ± 4.99 | 0.93
AFE | −0.49 ± 0.43 | −0.11 ± 0.23 | −0.03 ± 0.19 | 0.629 | 0.08 ± 0.28 | −0.21 ± 0.20 | −0.10 ± 0.27 | 0.70
AFE_EW | −0.09 ± 0.06 | 0.03 ± 0.03 | 0.01 ± 0.02 | 0.16 | −0.01 ± 0.04 | 0.02 ± 0.03 | 0.00 ± 0.04 | 0.76
AFE_WT | 1.75 ± 5.02 | 8.30 ± 5.47 | −3.23 ± 4.51 | 0.27 | −3.55 ± 6.60 | 3.72 ± 4.72 | 2.02 ± 6.40 | 0.67
EN270 | 1.79 ± 2.38 | 1.76 ± 1.30 | −0.99 ± 1.07 | 0.22 | 1.26 ± 1.57 | −0.51 ± 1.12 | 0.89 ± 1.52 | 0.59
EN360 | 1.98 ± 0.72a | 1.37 ± 0.39a | −0.51 ± 0.32b | <0.0001 | −0.24 ± 0.48a | 0.29 ± 0.34ab | 1.33 ± 0.46b | 0.04
EW270 | 0.11 ± 0.13 | −0.04 ± 0.07 | 0.04 ± 0.06 | 0.51 | 0.05 ± 0.08 | 0.04 ± 0.06 | −0.05 ± 0.08 | 0.63
EW360 | 0.08 ± 0.09 | −0.03 ± 0.05 | 0.03 ± 0.04 | 0.51 | 0.04 ± 0.06 | 0.03 ± 0.04 | −0.04 ± 0.06 | 0.62
PLS | −0.03 ± 0.26 | 0.04 ± 0.14 | −0.08 ± 0.12 | 0.80 | −0.24 ± 0.17 | 0.08 ± 0.12 | −0.03 ± 0.17 | 0.31
PR | 0.21 ± 1.22 | −0.09 ± 0.66 | −0.31 ± 0.55 | 0.91 | −1.33 ± 0.80 | 0.28 ± 0.57 | 0.06 ± 0.77 | 0.25

PCA and trait clustering

The characterization based on principal components indicates that growth, sexual maturity, cumulative egg production, and egg-laying persistency represent partially independent biological dimensions rather than a single productivity continuum. The first principal component consistently captured overall growth patterns in early body weight traits, suggesting that somatic growth constitutes a stable, genetically structured dimension largely distinct from reproductive timing and egg-laying performance. This pattern is consistent with previous quantitative genetic studies reporting limited overlap between growth-related traits and egg production or persistency [9, 38].

Previous multivariate analyses have shown that early and cumulative egg production load on separate principal components, with early production reflecting initial laying intensity and later components capturing variation associated with persistency across the laying cycle [10]. Similarly, Venturini et al. [39] reported that total egg output and post-peak production decline are represented by different principal components, indicating that egg number and egg-laying persistency can be statistically distinguished despite moderate correlations. More recent studies further suggest that persistency loads independently from body weight and production rate, reflecting potential trade-offs between sustained laying capacity and other economically important traits [18].

In the present study, the three-dimensional PCA score plot revealed a latent phenotypic pattern in which favorable persistency and higher cumulative egg production were distributed along a distinct production–persistency axis, while substantial dispersion remained across growth and maturity axes. This multiaxial structure is consistent with longitudinal and genomic evidence identifying egg-laying persistency as a distinct dimension contributing to late-cycle productivity and adaptability [14, 40, 41]. Together, these results support consideration of egg-laying persistency as a complementary trait to total egg number in breeding programs targeting long-term production stability. For example, incorporating persistency-related traits into multi-trait selection indices may help identify hens with more stable long-term laying performance rather than selecting solely for early cumulative egg number.

Effects of candidate gene polymorphisms on trait clustering

The PCA genotype scatter plots showed substantial overlap among genotypes for all candidate genes, indicating that growth, egg production, and egg-laying persistency are not dominated by single major loci. This pattern is consistent with a polygenic genetic architecture for economically important traits, as reported in previous longitudinal and quantitative genetic studies [40]. The limited genotype-specific separation observed in PCA space suggests that polymorphisms in DRD2, VIP, NPY, and MTNR1C exert modest, trait-dependent effects rather than defining distinct multivariate phenotypic profiles.

The absence of distinct genotype clustering may reflect the highly polygenic nature of these traits and the limited ability of PCA to detect subtle effects of individual loci relative to total phenotypic variation. This observation is consistent with genome-wide association studies showing that reproductive traits are typically influenced by numerous loci with small effects [42]. Despite the absence of clear phenotypic clustering, DRD2 and MTNR1C exhibited moderate polymorphism and balanced allele frequencies (Table-2), indicating that these loci retain genetic variation within the population. In contrast, deviations from the Hardy–Weinberg equilibrium observed for VIP and NPY may reflect population structure or historical selection pressure. These loci have been implicated in neuroendocrine pathways associated with reproductive regulation, suggesting potential biological relevance [13]. Overall, the present results are consistent with recent genomic studies indicating that candidate genes are more appropriately considered within multivariate or genomic selection frameworks rather than as single-gene predictors of performance [14].

Genetic parameter estimates

The heritability estimates in this study are largely consistent with previous reports in native and tropical chicken populations. The very high heritability for BW0 (h² = 0.72) is slightly higher than most published estimates for Thai native and indigenous chickens but follows the same pattern of stronger genetic determination at hatch and early life [6, 43]. As birds aged, heritability of body weight declined to moderate levels (0.24–0.35), which is consistent with estimates reported for Thai native, Korean native, Iranian Mazandaran, and Ethiopian Tilili chickens [7, 11, 16]. This reduction likely reflects increasing environmental and management influences and is particularly relevant under tropical conditions, where growth rate is constrained by heat stress and resource availability. Differences in inheritance rates across ages may result from increased environmental variation with advancing age [43, 44].

In Thai native chickens, Boonkum et al. [34] demonstrated that heat stress slows growth while maintaining moderate additive genetic variance, supporting the present observation that later growth traits are more strongly environmentally modulated. AFE_WT showed moderate heritability (h² = 0.47), consistent with estimates reported in Thai purebred and hybrid layers [45], supporting its potential relevance as a selection criterion. In contrast, the low heritability estimates for egg number (0.08–0.12) closely match those reported in both native and commercial chickens reared under tropical environments [10, 45]. Egg weight exhibited low-to-moderate heritability, consistent with reports indicating relatively stable genetic control across breeds [17]. Importantly, the heritability of PR (h² = 0.43) falls within the moderate range reported for persistency or post-peak stability in commercial layers, whereas the lower heritability of PLS suggests greater environmental sensitivity during the late laying period [12, 40].

Genetic correlations revealed biologically relevant trade-offs that distinguish Thai native chickens from intensively selected commercial layers. Positive genetic correlations between early body weight and AFE indicate that genetically heavier birds tend to mature later, consistent with previous findings in both Thai native and commercial populations [16, 45]. In contrast, egg persistency traits showed consistently negative genetic correlations with growth traits and cumulative egg number, indicating that alleles favoring rapid growth or high total egg output tend to be associated with reduced long-term laying stability. The strong negative genetic correlations between egg number traits (EN270 and EN360) and persistency traits (PR and PLS) demonstrate that, in Thai native chickens, high egg number does not necessarily correspond to stable or persistent egg production.

This pattern contrasts with many studies in commercial layers, where persistency and egg number are often favorably correlated under controlled environments [12, 40]. Previous studies using random regression and spline-based approaches in commercial and Thai native chickens generally reported moderate-to-high positive genetic correlations between egg number and persistency-related traits, particularly during post-peak production stages [46]. This contrast is consistent with emerging evidence that egg-laying persistency represents a distinct biological dimension linked to metabolic resilience and reproductive longevity rather than cumulative egg output alone [14]. Nevertheless, the positive residual correlations between egg number and persistency observed in this study suggest that improvements in nutrition, housing, and health management may simultaneously enhance both traits at the phenotypic level. Collectively, these results emphasize that sustainable genetic improvement in Thai native chickens may benefit from multi-trait selection indices that explicitly incorporate egg-laying persistency alongside egg number and growth traits, in combination with management strategies that mitigate environmental constraints such as heat stress [34, 45].

Candidate genes for selection in Thai native chickens

The candidate genes evaluated in this study (DRD2, VIP, NPY, and MTNR1C) were selected because they are known regulators of neuroendocrine pathways controlling reproduction, energy balance, and circadian rhythms in poultry. Ngu et al. [23] demonstrated that VIP plays a central role in regulating prolactin secretion and reproductive activity in birds, whereas Tu et al. [21] reported that dopaminergic signaling through DRD2 exerts inhibitory control over prolactin release. Ngu et al. [23] further showed that prolactin-related endocrine regulation primarily influences laying performance rather than somatic growth.

In the present study, no consistent associations were detected between polymorphisms in DRD2, VIP, NPY, or MTNR1C and EBVs for growth traits, AFE, or egg weight, indicating that these genes are unlikely to represent major genetic determinants of growth or sexual maturity. This pattern is consistent with the findings of Fu et al. [47], who reported in a genome-wide association study that growth and egg weight traits are highly polygenic and influenced by numerous loci of small effect rather than by single candidate genes.

In contrast, significant genotype-dependent differences were observed for EBVs of cumulative egg number, particularly EN360, suggesting a trait-specific contribution of the investigated genes to long-term egg production. The present findings for MTNR1C and NPY are consistent with previous studies reporting significant associations of neuroendocrine-related genes with cumulative egg production traits in chickens [26, 48]. Similar associations between melatonin receptor gene polymorphisms and egg production traits have also been reported in chickens and ducks [24, 25, 47]. Collectively, these findings support the role of DRD2, VIP, NPY, and MTNR1C as supplementary molecular markers for improving long-term egg yield in breeding programs. Because EBVs are shrinkage-based estimates derived from mixed models, candidate gene associations should be interpreted cautiously, as part of the observed variation may depend on the genetic evaluation model.

Although persistency definitions are simple ratios, their integration with PCA and candidate genes provides a practical framework readily applicable to other tropical indigenous breeds. These findings suggest that incorporating persistency-related EBVs into multi-trait selection strategies may help balance long-term productivity and resilience in indigenous chicken breeding programs. Several limitations should be considered when interpreting the present findings. Egg number traits showed relatively low heritability, and tropical environmental conditions may have increased environmental variance affecting reproductive performance. In addition, candidate gene associations were based on EBVs derived from mixed models, while deviations from Hardy–Weinberg equilibrium at the VIP and NPY loci may reflect selection history or population structure. Furthermore, the functional effects of the studied polymorphisms were not experimentally validated.