RESULTS

Summary statistics of growth performance and egg production traits

The descriptive statistics for body weight, egg production, and egg persistency traits are presented in Table-1. The mean BW0 of Thai indigenous chickens was 29.10 ± 3.35 g, with moderate variability (%CV = 11.51). Body weight increased progressively with age, reaching 255.23 ± 50.93 g at BW4, 735.36 ± 95.37 g at BW8, 1160.35 ± 113.65 g at BW12, and 1464.68 ± 127.97 g at BW16. The %CV declined with advancing age, from 19.96 at BW4 to 8.74 at BW16, indicating reduced relative variability in body weight as birds matured. Considerable phenotypic ranges were observed for all growth traits, particularly at later ages, with BW16 ranging from 1100 to 2182 g.

The mean AFE_WT was 2031.90 ± 210.78 g, while the mean AFE was 162.80 ± 15.17 days. AFE_EW averaged 36.06 ± 6.32 g, showing relatively high variability (%CV = 17.52). EN270 averaged 154.53 ± 28.76 eggs and increased to 181.46 ± 31.78 eggs at EN360. Both traits exhibited moderate-to-high variability, with %CV values of 18.61 and 17.52, respectively, and wide ranges across individuals. EW270 averaged 46.18 ± 4.39 g and increased slightly to 46.77 ± 4.82 g at EW360, with relatively low variability compared with egg number traits.

PR averaged 0.24 ± 0.24, with a %CV of 100.00 and values ranging from 0.00 to 1.82. Similarly, persistency of laying score (PLS) averaged 0.15 ± 0.11, with a high %CV of 72.64 and a maximum value of 0.58. These results indicate substantial individual variation in laying persistency during the later production period.

Traits | N | Mean | SD | %CV | Min | Max
Growth performance |  |  |  |  |  | 
BW0 (g) | 507 | 29.10 | 3.35 | 11.51 | 19.00 | 40.00
BW4 (g) | 507 | 255.23 | 50.93 | 19.96 | 115.00 | 406.00
BW8 (g) | 507 | 735.36 | 95.37 | 12.97 | 443.00 | 1004.00
BW12 (g) | 490 | 1160.35 | 113.65 | 9.79 | 796.00 | 1616.00
BW16 (g) | 484 | 1464.68 | 127.97 | 8.74 | 1100.00 | 2182.00
Egg production |  |  |  |  |  | 
AFE_WT (g) | 545 | 2031.90 | 210.78 | 10.37 | 1075.00 | 2652.00
AFE (d) | 545 | 162.80 | 15.17 | 9.32 | 130.00 | 211.00
AFE_EW (g) | 545 | 36.06 | 6.32 | 17.52 | 21.00 | 64.00
EN270 (egg) | 545 | 154.53 | 28.76 | 18.61 | 53.00 | 221.00
EN360 (egg) | 545 | 181.46 | 31.78 | 17.52 | 83.00 | 266.00
EW270 (g) | 427 | 46.18 | 4.39 | 9.52 | 32.00 | 59.00
EW360 (g) | 545 | 46.77 | 4.82 | 10.30 | 31.00 | 62.00
Egg persistency |  |  |  |  |  | 
ENearly | 545 | 153.97 | 28.31 | 18.39 | 64.00 | 221.00
ENlate | 545 | 27.45 | 20.77 | 75.67 | 0.00 | 110.00
PR | 545 | 0.24 | 0.24 | 100.00 | 0.00 | 1.82
PLS | 545 | 0.15 | 0.11 | 72.64 | 0.00 | 0.58

PCA and trait clustering

PCA was performed to evaluate multivariate relationships among growth performance, reproductive timing, egg production, and egg persistency traits (Figure-2). Bartlett’s test of sphericity indicated that the data were suitable for PCA (χ² = 6465.336, df = 91, p < 0.001). Based on an inspection of the scree plot and PC-based parallel analysis, four PCs were retained, explaining 71.7% of the total phenotypic variance. This multivariate structure highlights distinct biological clustering among growth, maturity, egg production, and persistency traits that have been less clearly characterized in Thai native chicken populations.

PC1 explained 22.6% of the variance and showed high loadings for body weight traits measured from hatch to 16 weeks of age (BW0–BW16). PC2 accounted for 19.6% of the variance and was primarily associated with AFE and the corresponding body weight and egg weight traits. PC3 explained 16.0% of the variance and was characterized by high loadings for egg number and egg weight traits at later stages of production. PC4 accounted for 13.4% of the variance and exhibited strong loadings for persistency-related traits (PR and PLS), along with a positive association with EN270. The PCA loading pattern indicated clear clustering of growth performance, maturity, egg production, and egg persistency traits across distinct components (Figure 2).

The three-dimensional PCA plot (Figure-3) showed separation of individuals along the Prin3 axis, corresponding primarily to egg production and persistency traits. Individuals with higher Prin3 scores exhibited higher values for persistency-related traits and cumulative egg production, whereas variation along Prin1 and Prin2 reflected differences in growth and maturity-related traits. A limited number of individuals displayed favorable scores across all three principal axes.

Effects of candidate gene polymorphisms on trait clustering

Figures 4–6 present PCA plots of all phenotypic traits, with individuals grouped by DRD2, MTNR1C, NPY, and VIP genotypes. Across all PCA planes (PC1–PC4), genotype classes showed wide dispersion and substantial overlap, with no distinct genotype-specific clustering or separation. Similar distribution patterns were observed for growth, sexual maturity, cumulative egg number, and egg persistency traits. Persistency-related components displayed variation largely independent of growth- and maturity-associated components across genotypic groups.

Overall, PCA patterns did not indicate discrete multivariate phenotypic profiles associated with any single polymorphism. The extensive overlap among genotype groups suggests that individual candidate genes had relatively small effects on overall phenotypic variation in Thai native chickens.

Table 2 presents the genotype and allele frequencies, polymorphism information content, and Hardy–Weinberg equilibrium test results for candidate genes in Thai native chickens. The allele frequencies of MTNR1C were relatively balanced between alleles. Polymorphism information content values for DRD2, VIP, MTNR1C, and NPY were moderate, ranging from 0.33 to 0.38, indicating moderate levels of genetic polymorphism within the population. In addition, the DRD2 and MTNR1C loci conformed to Hardy–Weinberg equilibrium (χ² < 3.84). In contrast, the VIP locus exhibited a strong deviation from equilibrium (χ² = 119.23) despite showing moderate polymorphism. The NPY locus showed a high frequency of the S allele (0.71) and deviation from Hardy–Weinberg equilibrium (χ² = 7.26).

Figure 2: (A) PCA correlation biplot network showing variable loadings of the first four rotated components. Green and red lines indicate positive and negative loadings, respectively, and line thickness reflects loading magnitude. The first rotated component is associated with growth performance traits (BW0–BW16), the second with sexual maturity and egg weight traits (AFE, AFE_WT, AFE_EW, EW270, and EW360), the third with egg persistency traits (PR and PLS), and the fourth with cumulative egg production traits (EN270 and EN360). (B) Scree plot of eigenvalues for phenotypic traits. Based on the elbow rule and Kaiser criterion (eigenvalue >1), four components were retained.

Figure 3: Three-dimensional PCA score plot showing the distribution of observations along Prin1 (growth), Prin2 (maturity), and Prin3 (egg production and persistency). The circled region indicates observations with high Prin3 scores, representing favorable egg-laying persistency and cumulative egg production. Axis dispersion suggests partial independence among growth, maturity, and egg production dimensions.

Genetic parameter estimates

Genetic parameter estimates derived from models incorporating candidate gene marker information are presented in Table 3. For growth performance traits, heritability estimates ranged from moderate-to-high, with the highest estimate observed for BW0 (h² = 0.72 ± 0.09). Post-hatch body weights exhibited moderate heritability, including BW4 (0.24 ± 0.08), BW8 (0.35 ± 0.10), BW12 (0.44 ± 0.10), and BW16 (0.35 ± 0.08).

Figure 4: PCA scatter plots showing the distribution of individuals across (A) PC1 and PC2 and (B) PC1 and PC3 based on all phenotypic traits, with individuals grouped by DRD2, MTNR1C, NPY, and VIP genotypes. Extensive overlap among genotype groups indicates that genotype-associated variation is small relative to total phenotypic variation across the PCs.

Figure 5: PCA scatter plots showing the distribution of individuals across (A) PC1 and PC4 and (B) PC2 and PC4 based on all phenotypic traits, with individuals grouped by DRD2, MTNR1C, NPY, and VIP genotypes. Extensive overlap among genotype groups indicates that genotype-associated variation is small relative to total phenotypic variation across the PCs.

For egg production traits, AFE_WT showed moderate heritability (0.47 ± 0.10), whereas AFE and AFE_EW exhibited lower heritability estimates of 0.22 ± 0.07 and 0.07 ± 0.05, respectively. Cumulative egg number showed low heritability at EN270 (0.08 ± 0.03) and EN360 (0.12 ± 0.05). Egg weight at later production stages exhibited low-to-moderate heritability, with estimates of 0.23 ± 0.08 for EW270 and 0.10 ± 0.07 for EW360.

Egg persistency traits exhibited low-to-moderate heritability estimates, with values of 0.43 ± 0.11 for PR and 0.20 ± 0.09 for PLS. The moderate heritability estimated for PR represents a potentially useful finding for persistency-related selection in Thai native chickens. Across most egg production traits, residual variance exceeded additive genetic variance, whereas a greater proportion of additive genetic variance was observed for early growth and persistency-related traits.

Genetic and residual correlations among traits

Genetic and residual correlations among growth performance traits, age and weight at first egg, egg production traits, and egg persistency traits are presented in Table 4. Genetic correlations among body weight traits measured at different ages (BW0–BW16) were positive and ranged from low to high magnitude (0.15–0.67).

Figure 6: PCA scatter plots showing the distribution of individuals across (A) PC2 and PC3 and (B) PC3 and PC4 based on all phenotypic traits, with individuals grouped by DRD2, MTNR1C, NPY, and VIP genotypes. Extensive overlap among genotype groups indicates that genotype-associated variation is small relative to total phenotypic variation across the PCs.

Gene | Genotype frequencies | Allele frequencies | χ² | PIC | He | p-value
DRD2 | CC = 0.41 | CT = 0.47 | TT = 0.12 | C = 0.65 | T = 0.36 | 0.55 | 0.35 | 0.45 | 0.866
VIP | II = 0.50 | ID = 0.24 | DD = 0.26 | I = 0.62 | D = 0.38 | 119.23 | 0.36 | 0.47 | <0.001
MTNR1C | AA = 0.25 | AG = 0.49 | GG = 0.26 | A = 0.50 | G = 0.51 | 0.42 | 0.38 | 0.50 | 0.899
NPY | LL = 0.11 | MM = 0.36 | SS = 0.53 | L = 0.29 | S = 0.71 | 7.26 | 0.33 | 0.41 | 0.172

Traits | Additive variance | Residual variance | Heritability (h²) | SE
Growth performance |  |  |  | 
BW0 (g) | 8.0 | 3.1 | 0.72 | 0.09
BW4 (g) | 280.1 | 898.3 | 0.24 | 0.08
BW8 (g) | 1743.0 | 3205.9 | 0.35 | 0.10
BW12 (g) | 4556.1 | 5686.5 | 0.44 | 0.10
BW16 (g) | 5499.4 | 10141.0 | 0.35 | 0.08
Egg production |  |  |  | 
AFE_WT (g) | 16196.0 | 18066.0 | 0.47 | 0.10
AFE (d) | 29.7 | 107.9 | 0.22 | 0.07
AFE_EW (g) | 2.3 | 30.6 | 0.07 | 0.05
EN270 (egg) | 70.1 | 780.6 | 0.08 | 0.03
EN360 (egg) | 113.7 | 627.6 | 0.12 | 0.05
EW270 (g) | 2.8 | 9.4 | 0.23 | 0.08
EW360 (g) | 1.3 | 12.0 | 0.10 | 0.07
Egg persistency |  |  |  | 
PR | 207.6 | 278.9 | 0.43 | 0.11
PLS | 17.9 | 70.7 | 0.20 | 0.09

Early body weight traits showed positive genetic correlations with AFE, with estimates of 0.47 for BW4 and 0.22 for BW8. In contrast, BW4 and BW8 exhibited negative genetic correlations with EN360, ranging from −0.23 to −0.43.

Egg persistency traits showed negative genetic correlations with body weight from 4 to 16 weeks of age (BW4–BW16). Genetic correlations between BW4–BW16 and PR ranged from −0.54 to −0.39, whereas those with PLS ranged from −0.36 to −0.33. EN270 and EN360 also showed strong negative genetic correlations with egg persistency traits. Genetic correlations between EN270 and PR and PLS were −0.88 and −0.70, respectively, whereas corresponding correlations between EN360 and PR and PLS were −0.82 and −0.87. These strong negative genetic correlations indicated unfavorable relationships between cumulative egg number and persistency traits in slow-growing Thai native chickens.

Residual correlations between EN360 and egg persistency traits were positive, ranging from 0.57 to 0.80. Residual correlations among growth, egg production, and persistency traits generally differed in magnitude and direction from their corresponding genetic correlations.

Traits | BW0 | BW4 | BW8 | BW12 | BW16 | AFE_WT | AFE | AFE_EW | EW270 | EN270 | EW360 | EN360 | PR | PLS
BW0 | — | 0.39 | 0.38 | 0.26 | 0.54 | 0.19 | −0.13 | −0.15 | −0.11 | 0.05 | −0.21 | −0.06 | 0.20 | 0.52
BW4 | 0.35 | — | 0.48 | 0.67 | 0.55 | 0.35 | 0.47 | −0.40 | 0.28 | 0.08 | 0.05 | −0.23 | −0.39 | −0.33
BW8 | 0.55 | 0.76 | — | 0.15 | 0.42 | −0.31 | 0.22 | −0.52 | −0.12 | −0.14 | −0.24 | −0.43 | −0.51 | −0.35
BW12 | 0.38 | 0.35 | 0.77 | — | 0.36 | 0.75 | 0.01 | 0.05 | 0.28 | 0.23 | 0.18 | 0.22 | −0.50 | −0.35
BW16 | 0.46 | 0.41 | 0.53 | 0.64 | — | 0.21 | −0.08 | −0.52 | 0.24 | 0.48 | 0.23 | 0.05 | −0.54 | −0.36
AFE_WT | 0.22 | 0.20 | 0.21 | 0.15 | 0.27 | — | −0.06 | 0.17 | 0.17 | 0.40 | 0.15 | 0.57 | 0.25 | 0.20
AFE | 0.21 | −0.31 | −0.03 | 0.03 | 0.03 | −0.73 | — | −0.20 | 0.17 | −0.20 | 0.01 | −0.35 | 0.25 | 0.29
AFE_EW | 0.53 | 0.17 | 0.20 | 0.05 | 0.18 | 0.24 | −0.05 | — | 0.28 | −0.49 | 0.32 | −0.07 | 0.64 | −0.01
EW270 | 0.44 | 0.00 | 0.11 | 0.15 | 0.10 | −0.15 | 0.47 | −0.06 | — | −0.09 | 0.92 | −0.29 | 0.06 | 0.06
EN270 | −0.21 | 0.12 | 0.00 | −0.12 | −0.01 | −0.30 | 0.36 | −0.05 | 0.04 | — | −0.03 | 0.80 | −0.88 | −0.70
EW360 | 0.44 | 0.09 | 0.14 | 0.09 | −0.01 | −0.08 | 0.37 | −0.06 | 0.85 | 0.00 | — | −0.17 | −0.07 | −0.07
EN360 | −0.03 | 0.29 | 0.06 | −0.08 | 0.03 | −0.06 | 0.05 | −0.18 | 0.00 | 0.71 | −0.04 | — | −0.82 | −0.87
PR | −0.02 | 0.03 | 0.06 | 0.09 | 0.07 | 0.03 | −0.20 | −0.75 | 0.00 | −0.12 | 0.38 | 0.80 | — | 0.92
PLS | 0.00 | 0.38 | 0.26 | 0.30 | 0.30 | 0.16 | 0.16 | 0.00 | 0.00 | 0.09 | 0.09 | 0.57 | 0.92 | —

Associations between candidate genes and traits

Table 5 presents the associations of DRD2 and VIP polymorphisms with EBVs for growth performance, egg production, and persistency traits in Thai native chickens. Polymorphisms in DRD2 and VIP were significantly associated with EBVs for EN270 and EN360 (p < 0.05).

Table 6 presents the associations of NPY and MTNR1C polymorphisms with EBVs for growth performance, egg production, and persistency traits in Thai native chickens. Polymorphisms in NPY and MTNR1C were significantly associated with EBVs for EN360 (p < 0.05).

Chickens carrying the DRD2 CC genotype showed significantly higher EBVs for EN360 than those with the TT genotype (1.61 vs. −4.14). Similarly, the VIP DD genotype was associated with superior egg production performance compared with the II genotype, particularly for EN360 (2.84 vs. −0.64) and EN270 (4.22 vs. −1.10). In contrast, the NPY SS and MTNR1C AA genotypes were associated with lower egg production performance. However, no significant associations were detected between these candidate genes and growth or laying persistency traits.