Data were analyzed for the 2 sets. First set was only for the first parity and the second set was all the parities of animals over lifetime. Table 1 shows the summary statistics for lactation traits across lifetime. For Jamunapari goat, the least squares means for MY90, MY140, and TMY were 78.62 ± 0.44, 113.45 ± 0.59, and 121.05 ± 0.73 kg, respectively. The average lactation length was 175.33 ± 0.68 days, which was comparatively longer than that of other Indian breeds. Jamunapari goats were in production until the seventh parity; nevertheless, some goats persisted to be productive until the eleventh parity. Over the years, parity and period of birth had a significant (P ≤ 0.01) effect on milk traits. The season of birth exhibited a significant (P ≤ 0.01) effect on MY90 and MY140. Does with multiple kids produced more milk than those with single kid. Higher estimates than the current study for milk traits were reported by Valencia et al. (2007) in Saanen goat (800kg, 338Kg and 285 days for TMY, MY120 and LL, respectively) and Kominakis et al. (2000) in Skopelos dairy goat of Greece (164.7kg and 239.2Kg for MY90 and TMY, respectively), owing to breed differences.
Genetic parameter estimates for lifetime lactation traits
Table 2 displays the parameter estimates considering all parities for milk yield traits and lactation length. The animal model 3 (out of model 3, 4 and 5) fitting the permanent environment was the most acceptable model for MY90, MY140 and TMY. For LL, the model 5 was most appropriate (Table 2). Since milk yield is greatly influenced by the animal's permanent environment, model 3 appeared to be the most appropriate model for the MY90, MY140, and TMY. MY90, MY140, and TMY had estimates of direct additive heritability as 0.10 ± 0.03, 0.08 ± 0.03 and 0.12 ± 0.02, respectively. These estimates for lifetime milk production traits were low to moderate. For LL, the direct heritability was 0.06 ± 0.02, which was low. The maternal genetic effect was also important for LL, although maternal heritability was low (0.04 ± 0.02). Because of the large datasets, heritability estimates for different traits were considerably different from zero, with small standard errors (0.02–0.03).
Table 2
Estimates of (co) variance component and genetic parameters for lifetime lactation traits by univariate analyses in Jamunapari goat
Trait | MY90 | MY140 | TMY | LL |
σ2a | 65.42 ±19.68 | 92.10 ± 33.86 | 267.33 ± 51.81 | 106.82 ± 32.66 |
σ2m | | | | 72.70 ± 35.52 |
σ2c | 72.58 ±18.88 | 116.97 ± 34.57 | 155.05 ± 39.66 | 12.03 ± 38.74 |
σ2e | 502.66 ± 13.67 | 931.38 ± 28.21 | 1844.34 ± 48.59 | 1693.43 ± 43.76 |
σ 2p | 640.66 ± 15.71 | 1141.35 ± 28.88 | 2266.72 ± 54.66 | 1944.90 ± 52.97 |
h 2 | 0.10 ± 0.03 | 0.08 ± 0.03 | 0.12 ± 0.02 | 0.06 ± 0.02 |
c2 | 0.11 ± 0.03 | 0.10 ± 0.03 | 0.07 ± 0.02 | 0.006 ± 0.02 |
m2 | | | | 0.04 ± 0.02 |
rep | 0.22 ± 0.02 | 0.18 ± 0.02 | 0.19 ± 0.02 | 0.17 ± 0.59 |
Log L | -14604.636 | -13032.170 | -17542.462 | -17231.809 |
MY90 = 90 days milk yield; MY140 = 140 days milk yield; TMY = total milk yield; LL = lactation length; σ2a = additive direct genetic variance; σ2m = maternal genetic variance; σ2c = animal permanent environmental variance; σ2e = residual variance; σ2p = phenotypic variance; h2 is heritability; m2 is σ2m/σ2p; c2 is σ2c/σ2p; rep = repeatability; log L is log likelihood. |
The estimates of direct additive heritability in the current study for MY90, MY140, TMY and LL are pursuant with previous findings in literature (Kominakis et al., 2000; Valencia et al., 2007; Weppert and Hayes, 2004; Torrens-Vazquez et al., 2009). In Greece, the heritability of MY90 and total milk production in the Skopelos breed was 0.15 and 0.14, respectively (Kominakis et al 2000). Weppert and Hayes (2004) reported heritability estimates of 0.13 for milk yield in the Alpine, Toggenburg, Saanen, and Nubian breeds. In Mexico, the heritability for total milk production, MY120 (milk yield at 120 day), and LL were 0.22, 0.12, and 0.04, respectively (Valencia et al., 2007), and 0.17 for milk yield in Saanen goats (Torres-Vázquez et al., 2009). However, higher estimates than current study were reported by (Kennedy et al., 1982; Boichard et al., 1989; Muller et al., 2002). Kennedy et al. (1982) reported heritability of 0.68, 0.61, 0.54, and 0.30 for milk yield in the Alpines, Saanens, Toggenburgs, and Nubian goat breeds respectively. Boichard et al. (1989) obtained heritability estimates of 0.29–0.31for milk yield in French Alpine and Saanen primiparous goats. Muller et al. (2002) estimated the heritability of milk yield in South African Saanens to be 0.23.
The variance component indicates that almost all milk traits are largely controlled by environmental factors, with the σ2a contributing little to total variance. Animal permanent environment influence (c2) played a significant role in milk yield attributes, owing to repeated nature of these traits in each parity. The estimates for c2 for MY90, MY140, TMY and LL were 0.11 ± 0.03, 0.10 ± 0.03, 0.07 ± 0.02 and 0.006 ± 0.02, respectively. We find that for the 90 and 140 days milk yield, the individual permanent environment remains significant source and contributes for more than 10% of total variance due to similar intrauterine environment and nursing for kids born to same mother over lifetime. For milk yield traits, the additive maternal genetic effects were absent, and only for LL the m2 was observed which was contributing for 4% of total variance. Maternal genetic variations for lactation milk yield have been found to be minimal and non-significant across literature (Van Vleck and Bradford, 1966; Reed and Van Vleck, 1987; Kirkpatrick and Dentine, 1988; Schutz et al., 1992; Albuquerque et al., 1995; Khattab et al., 2005) in dairy cattle and in dairy goats by Weppert and Hayes, (2004).
Estimates of repeatability set upper bounds on heritability estimates. The milk production traits had modest repeatability estimates of 0.22, 0.18, 0.19 and 0.17 for MY90, MY140, TMY and LL, respectively. Estimates of moderate repeatability suggest that successful selection for that trait is possible. Our estimates of repeatability were lower than those reported by other researchers (Montaldo et al., 1982; Constantinou et al., 1985; Ilahi et al., 1998; Bagnika and Lukaszewics, 1999; Van der Linde, 2002; Valencia et al., 2007; Torres-Vázquez et al., 2009). Montaldo et al. (1982) estimated 0.59 and 0.23 repeatabilities for TMY and LL, respectively, in Mexico. Constantinou et al. (1985) reported repeatabilities of 0.43, 0.36, 0.39, and 0.08 for MY90, MY150 (milk yield at 150 day), TMY, and LL, respectively in Damascus goat. Ilahi et al. (1998) computed a repeatability of 0.53 for milk yield of French Alpines. Bagnika and Lukaszewics (1999) reported a repeatability of 0.42 for milk yield in polish goat, while Van der Linde (2002) obtained a value of 0.55 for the same trait in the dairy goat of Netherlands. Valencia et al. (2007) reported repeatability estimates of 0.40, 0.27 and 0.11 for TMY, M120 and LL, respectively in Saanen goat. Torres-Vázquez et al. (2009) observed a repeatability of 0.43 for milk yield in the Mexican Saanen goat. Valencia (1992) reported lower repeatability estimates for TMY and LL (0.12 and 0.00) in Mexican goats than the current investigation.
Genetic analysis revealed use of 90-day milk yield as a suitable criterion of selection for further improvement
Table 3 shows the parameter estimates for milk yield traits and lactation length for first parity. Among models 1 and 2, the animal model 1 with its own additive effect was the most appropriate model for MY90 and MY140, TMY and LL. Estimates of direct heritability for MY90, MY140 and TMY were 0.26 ± 0.05, 0.26 ± 0.06 and 0.25 ± 0.06, respectively, which were moderate to high and indicate sufficient scope for selection, if further improvement in traits is desired. For LL, the estimate of h2 was 0.03 ± 0.04, which was low and defines that the LL cannot be further improved. Similar within range estimates for direct additive heritability for milk yield traits as compared to present study were reported earlier (Belichen et al., 1999; Montaldo et al., 2010; Rupp et al., 2011). Belichen et al. (1999) reported heritability estimates of 0.34 and 0.32 for milk yield in Alpine and seamen goat, respectively. Weppert and Hayes, (2004) obtained heritability for milk yield as 0.36, 0.38, 0.35 and 0.36 in Alpine, LaMancha, Nubian, Saanen and Toggenberg goats, respectively. Montaldo et al. (2010) found heritability for milk yield ranged from 0.35 to 0.38 in Alpine, LaMancha, Nubian, Saanen, and Toggenberg goat’s breeds. Rupp et al. (2011) observed heritability for milk yield of 0.30 and 0.34 in Alpine and seamen, respectively. However, lower estimates for heritability (0.13) than current study were reported by Weppert and Hayes (2004) in Alpine, Saanen, Toggenberg and Nubian breeds of goats. Moderate h2 estimates for first lactation yield traits suggested the presence of significant additive genetic variance for these traits, indicating the prospect of selection for future genetic progress in these traits.
Table 3
Estimates of (co) variance component and genetic parameters for first lactation traits in Jamunapari goat
Trait | MY90 | MY140 | TMY | LL |
σ2a | 140.07 ± 30.79 | 148.55 ± 30.08 | 465.20 ± 110.78 | 57.05 ± 70.76 |
σ 2p | 534.60 ± 21.78 | 945.25 ± 41.82 | 1876.99 ± 76.18 | 1740.09 ± 67.85 |
σ 2e | 394.53 ± 28.69 | 701.24 ± 56.59 | 1411.80 ± 103.60 | 1683.04 ± 93.83 |
h 2 | 0.26 ± 0.05 | 0.26 ± 0.06 | 0.25 ± 0.06 | 0.03 ± 0.04 |
Log L | -4758.420 | -4303.180 | -5631.986 | -5601.328 |
MY90 = 90 days milk yield, MY140 = 140 days milk yield, TMY = total milk yield, LL = lactation length,σ2a = additive direct genetic variance, σ2e = residual variance and σ2p = phenotypic variance; h2 is heritability; log L is log likelihood |
The genetic correlation estimates of MY90 with MY140 and TMY were 0.97 ± 0.01 and 0.91 ± 0.05, respectively (Fig. 1). The genetic correlation of MY140D with TMY was 0.98 ± 0.02. The estimates of phenotypic correlation between these three traits were also high and significant (0.86 to 0.97). In accordance with our findings, higher genetic and phenotypic correlations among lactation traits have also been reported in goats (Kominakis et al., 2000, Valencia, 2007; Roy and Mandal, 2010; Mucha et al., 2014). Results indicated that MY90 is the most suitable trait for early selection of the animals for milk yield owing to very high genetic correlation of MY90 with later expressed traits for milk production in Jamunapari goats. Hence, we recommend use of MY90 as the trait for early selection of animals.