Selection of Bali cattle based on birth weight and calving interval records at West Nusa Tenggara Province of Indonesia

Birth weight and calving interval are included of productivity traits that can be increased by selection program. However, the standard of desirable birth weight in cattle during the selection program is important to prevent dystocia incident risk. This study was aimed to select Bali cattle (Bos javanicus) based on Estimated Breeding Value of birth weight (EBVBW) and Most Probable Producing Ability of birth weight (MPPABW) and calving interval (MPPACI). Total of 758 records data of BW were collected from Lombok and Sumbawa islands, West Nusa Tenggara Province. Research showed the average of BW in Bali calves were 15.69±1.70 kg (Lombok) and 13.49±1.89 (Sumbawa). The average of CI in Bali cows at both islands were about 385 days. In addition, the heritability (h2) values of BW in both islands was about 0.90. The repeatability (r) values of BW in both islands were about 0.30. Meanwhile, the r Selection of Bali Cattle Based on Birth Weight and Calving Interval (S. Said et al.) 15 value of CI in Sumbawa island was 0.39. The highest of EBVBW for sire was +4.25 kg by bull’s ID: 0838 (Sumbawa). Meanwhile, the highest of EBVBW for calves was +6.07 kg by calf’s ID: 0917 (Sumbawa). The highest of MPPABW was +2.67 kg by cow’s ID: 0872 (Sumbawa). The lowest of MPPACI was -25.70 days by cow’s ID: 02076 (Lombok).


INTRODUCTION
Bali cattle (Bos javanicus) is one of Indonesian native cattle that adapted well in Bali island. Bali cattle were declared as the Indonesian native cattle since year 2010 through decision of Indonesian Ministry of Agriculture No: 325/Kpts/OT.140/1/2010. Kaswati et al. (2013) reported that the average of weaning weight (205 days of age) and yearling weight (365 days of age) in Bali cattle at the breeding station (Bali island) were 88.59±16.15 kg and 131.12±25.50 kg respectively. In addition, Priyanto et al. (2019) reported that the average of slaughter weight, carcass weight and dressing percentage in Bali cattle were 275.56±61.93 kg, 141.04±35.61 kg and 50.95±3.49% respectively. Recently, Bali cattle was adapted well in many islands of Indonesia. Mansur et al. (2016) reported that the average of adult weight in Bali cattle at Barru Regency (Sulawesi island) were 158.63±34.27 kg (male) and 185.72±46.37 kg (female) Soekardono et al. (2009) reported that the average of body weight in Bali cows (about 1.5 -2.0 years of age) were 157.00±20.59 kg (Lombok island) and 172.93±20.64 kg (Sumbawa island). In contrast, Samberi et al. (2010) reported that the reproductive efficiency (RE) in Bali cattle at Yapen archipelago (Papua Province) was 88.38% and included low category (RE < 100%).
Selection program to Bali cattle in the outsite of Bali island is important to increase the growth traits in cattle. Selection in the livestock can be performed with conventional and molecular methods. The conventional selection in the cattle can be obtain using Estimated Breeding Value (EBV) and Most Probable Producing Ability (MPPA) analysis. Unfortunately, the study to select Bali cattle in the outsite of Bali island with conventional method is not reported. Selection with conventional method is very complicated because this method need a large number of performance records and pedigree information. Therefore, most of the conventional selection study in Bali cattle were applied in the breeding station at Bali island.
This study was carried out to select Bali cattle in Lombok and Sumbawa islands with conventional selection method. In this study, two records data of birth weight and calving interval were used to select Bali cattle in West Nusa Tenggara (WNT) province. Previous studies reported that BW of calves can be used as the selection criteria to select the best cattle (Bakir et al., 2004;Putra et al., 2015;Said et al., 2016). Moreover, the genetic correlation (rg) between birth and weaning weights in cattle were included of moderate to high categories (rg > 0.50) such as Brangus (0.78), Fogera (0.60) and 0.87 for Sumba Ongole (Neser et al., 2012;Bekele et al., 2016;Putra et al., 2018). Hence, the birth weight can be used as the selection criteria to increase weaning weight. Moreover, the standard of desirable birth weight is important to prevent dystocia incident risk in cows. Meanwhile, the calving interval can be used as the selection criteria to obtain the cows with good reproductive trait. However, this study is important as the first conventional selection study in the Bali cattle outside of Bali island. In addition the results of this study can be used as the early information to select Bali cattle in WNT province with conventional method in the future.

Place and Samples
The records data of birth weight in Bali cattle from year 2012 to 2018 were collected from Bureau of Artificial Inseminantion (BIBD Banyumulek) and Bureau of Breeding and Forages Center (BPT-HPT Serading). In addition, both institutions of BIBD Banyumulek and BPT-HPT Serading were located at Lombok and Sumbawa islands respectively ( Figure 1). Therefore, both islands were included the main area of West Nusa Tenggara (WNT) province. The WNT province located between longitude 115° 46' -119° 5' E and latitude 8° 10' -9° 5' S about 11 to 3,726 m above the sea level. The maximum temperature in WNT province ranged between 31.1°C -33.0°C and minimum temperature ranged between 22.8°C -24.7°C. The humidity in WNT province about 79 to 85% and rainfall average occuring 1212 to 1800 mm/year. Amount of 758 records data of birth weight (BW) in Bali cattle from BIBD Banyumulek (400 records) and BPT-HPT Serading (358 records) were used in this study. Birth weight measurement was taken from each calf with spring weighing scales.

Management of Animals
The Bali cattle in BIBD Banyumulek (Lombok island) were managed with semiintensive management system in the farmer groups farm. The feed ration containing of king grass (Pennisetum purpuphoides), nutgrass (Cyperus rotundus), corncob and rice bran. Feed ration of corncob (4 kg), rice bran (6 kg). Thus, king grass (10 kg/head/day), corncob (5 kg/head/day), field grass (8 kg/head/day), and commercial concentrate (3 kg/head/day) were given to the cattle. The water was given by ad libitum and pregnancy and health examination were performed every month. The artificial insemination (AI) method was performed in this farm using Bali straw from BIBD Banyumulek. Therefore, the Bali cattle at BPT-HPT Serading (Sumbawa island) were kept by ranching management system. Every morning (09.00) to evening (16.00) cattle were managed in the pasture. The feed ration containing of elephant grass (Pennisetum purpureum), star grass (Cynodon plectostachyus) and rice bran. Feed ration of rice brand (1 kg/head/day) and concentrate (0.5 kg/head/day) were given to the cattle before managed in the pasture. The concentrate containing of water content (44.34%), dry matter (55.66%), crude protein (15.15%), TDN (80.07%), fat (13.15%) and rough fiber (14.26%). The natural mating method was performed in this breeding station using Bali bull. The herd was consisted of 1 bull and about 50 cows and every year the bull was exchanged. The water was given by ad libitum in the pasture and pregnancy and health examination were performed every month.

Standardization Data
Data of body weight (BW) in calves were standardized or corrected based on Hardjosubroto (1994) as follows: where CF Sex is the correction factor of body The calculaion of CF Sex only used for female calves and important to reduce the data variation from sex effect.

Data Analysis
Data of BW were analyzed to describe the effect of sex, year, parity and season. Data of sex, year and parity were collected from records data of herds book. Thus the season of birth was classified into two categories of dry (April -September) and rainy (October -March). Thus, the data were analyzed using General Linear Model (GLM) with formula according to Steel and Torrie (1980) as follows: Y ijkl = μ + S i + T j + P k + M l + E ijkl Where Y ijk is the observation of trait; μ is the common mean; S i is the effect of i th sex of calf; T j is the effect of j th year; P k is the effect of k th parity; M l is the effect of l th season; and E ijk is the experimental error.
The heritability (h 2 ) value was analyzed using paternal halfshib correlation model through analysis of variance (ANOVA) method with a mathematical model according to Becker (1992) as follows: Where Y i is the observation of trait; μ is the common mean; ɑ i is the effect of i th sire; and e i is the experimental error. Therefore, the h 2 value was estimated using formula according to Becker (1992) as follows: Where h 2 is the heritability value; σ S 2 is the variance component of sire; σ W 2 is the variance component of individu; SE(h 2 ) is the standard error of heritability; S is the number of sire; n i is the number of progeny per sire; and N is the total number of progeny. The repeatability (r) value was analyzed using interclass correlation method based on two records of birth weight with a mathematical model according to Warwick et al. (1990) as follows: Where r is the repeatability; X 1 is the record of trait 1; X 2 is the record of trait 2; and N is the number of data pairs.
The relative estimated breeding value (EBV) and Most Probable Producing Ability (MPPA) of birth weight were analyzed using mathematics formula according to Hardjosubroto (1994) as follows: Where EBV Sire is the estimated breeding value of sire; EBV Prog is the estimated breeding value of progeny; MPPA is the most probable producing ability of cow; n is the number of progeny; h 2 is the heritability value; r is the repeatability value; P Prog bar is the mean of trait in progeny; P pop bar is the mean of trait in population; and P Ind bar is the trait of individu.
The relative accuration (RA) calculation was performed in this study to measure the accuracy of selection for bulls using mathematical formula according to Hardjosubroto (1994) as follows: t = Rh 2 + c Where RA is the relative accuracy; n is the number of progeny per sire; R is 0.25 (halfsib correlation); c is 0.00 (no correlation assumed) or 0.10 (correlation assumed).

Birth Weight and Calving Interval
The BW of Bali calves in Lombok was higher than Sumbawa island (P<0.05) as presented in Table 1. The BW in male calves were higher than female calves at both islands.The BW of calves in Lombok island at year 2017 to 2018 were about 16.00 kg and higher than year 2013 to 2015 (about 15.00 kg). In this study, records data of BW in Lombok island for year 2012 and 2016 were not available. Meanwhile, the highest of BW in calves at Sumbawa island was showed at year 2012 and 2015 (about 15.00 kg). Commonly, the CI of calves that born in dry and rainy seasons were similar in both islands (about 385 days). According to the records data, most of calves in Lombok and Sumbawa islands were born in season of rainy (62%) and dry (92%) respectively. According to Table 1, three factors of sex, year and parity were affected to BW in both island. Meanwhile, factor of season was affected to BW in Sumbawa island.

Heritability and Repeatability
The h 2 value of BW in both islands were about 0.90 as presented in Table 2. Meanwhile, the r value of BW in both islands were about 0.30. Despite, the r value of CI in Sumbawa island was 0.39. In this study, the estimation of r value for CI in Lombok island was not performed because of no parallel parity records data. The SE values of h 2 were showed lower than h 2 value in both islands.

Estimated Breeding Value and Most Probable Producing Ability
The EBV BW in Bali bull at Lombok and Sumbawa islands were presented in Table 3 and  Table 4 respectively. Total records data of BW that used for h 2 and r estimations were 633 records and 120 records respectively. Thus, the h 2 value in this study was obtain using dataset from 13 bulls (Lombok island) and 7 bulls (Sumbawa island). Selection of Bali bulls in this study based on the two parameters of EBV Sire and RA. Bali calves were selected based on EBV prog for BW. Bali cows were selected by MPPA BW and MPPA CI .
The highest of relative EBV BW for sire at Lombok and Sumbawa islands were +3.94 kg (bull's ID: 11509d004) and +4.25 kg (bull's ID:  (Table 5). Overall the EBV BW for sire and calves in Lombok island were higher than Sumbawa island as show in Figure 2. Moreover, the highest of MPPA BW in Lombok and Sumbawa islands were +2.28 kg (cow's ID: 02009) and +2.67 kg (cow's ID: 0872) respectively (Table 6). In addition, the lowest MPPA CI in Lombok and Sumbawa islands were -25.70 days (cow's ID: 02076) and -26.59 (cow's ID: 0583) respectively (Table 7). Overall the MPPA BW in Lombok island was higher than Sumbawa island as shows in Figure 3. In addition, overall the MPPA CI in both island were similar.

DISCUSSIONS
The different of BW of Bali calves between Lombok and Sumbawa islands can be caused by different management system. The semi-intensive management and AI method can be caused the BW in Lombok island was higher than in Sumbawa island. Furthermore, the straws that used for AI were produced from selected bulls by BIBD Banyumulek. Tavares et al. (2012) reported that the BW of Bali calves in the breeding station at Bali island (BPTU-HPT Denpasar) were 18.37±1.65 kg (males) and 18.27±120 kg (females). Kaswati et al. (2013) reported that the corrected BW of Bali cattle in the BPTU-HPT Denpasar was 17.80±1.08 kg. Despite, Pemayun et al. (2014) reported that BW of Bali calves in the south Bali area were 15.62 kg (traditional system) and 16.94 kg (three strata forage system). Prasojo et al. (2010) reported that overall the BW of Bali calves in Gianyar, Bali and Badung (Bali island) was 18.40±1.60 kg. Gunawan and Jakaria (2011) reported that two factors of year and season were affected of BW in Bali calves at BPTU-HPT Denpasar. Factor of sex, year, parity and season also affect to BW in other Indonesian Ongole cross cattle (Hartati et al., 2015). Meanwhile, Putra et al. (2018) reported that two factors of year and season were affected of BW in Sumba Ongole cattle. According to the previous studies, the BW of Bali calves in Bali island was highest than in two islands at WNT province. The intensive selection program with good breeding practices for Bali cattle in BPTU-HPT Denpasar was caused the highly of BW in Bali island. Siswanto et al. (2013) reported that the CI of Bali cattle in BPTU-HPT Denpasar was 350.46±27.98 days and lower than in WNT province. In addition, Pemayun et al. (2014) reported that CI of Bali cattle in the south Bali area were 402 days (traditional system) and 384 days (three strata forage system). Soekardono et al. (2009) reported that the CI of Bali cattle in Lombok Barat (Lombok island) and Dompu (Sumbawa island) were 457.80 days and 470.10 days respectively. Despite, the CI of Bali cattle in Yapen archipelago (Papua) and Banyuasin (Sumatera island) were 410.40 days and 428.10 days respectively (Samberi et al., 2010;Susanti et al., 2015) and higher than in this study. The CI of Bali cattle in this study were similar to the Bali cattle in the south Bali area that managed with three strata forage system.
The h 2 value of BW in Bali cattle at WNT province was included of high category (h 2 >0.30) with low of SE value. High category of h 2 value for BW was showed in Red Chittagong (0.50), Horro (0.62), Nellore (0.37), Bali (0.85), Brahman  (Afroz et al., 2011;Abera et al., 2013;Regatieri et al., 2012;Kaswati et al., 2013;Rakwadi et al., 2014;Rahman et al., 2015;Putra et al., 2018). Moderate of h 2 value (0.10 < h 2 < 0.30) for BW was showed in Vrindavani (0.29), Tuli (0.21), Charolais (0.23) and Ongole cross (0.28) cattle (Singh et al., 2010;Rakwadi et al., 2014;Vostry et al., 2015;Hartati et al., 2015). Low of h 2 value (h 2 <0.10) for BW was showed in Aceh (0.06) and Fogera (0.06) cattle (Sari et al., 2016;Bekele et al., 2016). High h 2 value of BW suggested that selection based on this trait will be effective to increase gain of BW. The variation of h 2 value in this study compared to previous studies might be due to different of Selection of Bali Cattle Based on Birth Weight and Calving Interval (S. Said et al.) 21  (Hossain et al., 2018). The SE value in this study was lower than h 2 value and can be suggested that the h 2 estimation in this study was accurate. Furthermore, the h 2 value of BW in this study can be used as the one of the technical coefficient in the conventional selection program of Bali cattle at WNT province in the future.
The r values for CI in this study was included of high category (r>0.30). Recently, the new study to report the r value for CI of Bali or native cattle in Indonesia are not reported.
However, previous studies reported that high r value for CI were showed in Thari (0.35) and Kedah-Kelantan (0.34) and White Fulani (0.43) cattle (Sivarajasingam, 1975). The moderate r value (0.10 < r < 0.30) was showed in Tunisian Holstein (0.15), Ethiopian Holstein (0.23) and Mpwapwa (0.10) cattle (M'Hamdi et al., 2011;Ayalew et al., 2017;Chawala et al., 2017). The low r value (r < 0.10) for CI were showed in Japanese Black (0.09), crossbred beef (0.04),   (Oyama et al., 2002;Silva et al., 2015;Rahbar et al., 2016;Montaldo et al., 2017;Ersöz and Ertuğrul, 2018;Brazakova et al., 2019). High r value in this study can be suggested that overall cows had high repetition value (probability) to perform similar records data of BW and CI traits. Therefore, selection of Bali cows can be performed based on two records data of both traits. EBV BW and MPP ABW values of Bali cattle in this study were higher than reported in the Aceh bull (+0.42 kg), calf (-0.02 kg) and cow (+0.001 kg) in BPTU-HPT Indrapuri . Hilalah et al. (2018) reported that the highest EBV BW of Bali calf in BPTU-HPT Denpasar was +0.46 kg and lower than in this study. Intaratham   et al. (2008) reported that the highest EBV BW in Northeastern Thai indigenous cattle was reached of +74 kg and lower than in this study (+6.07 kg). In addition, Atil et al. (2005) reported that the EBV BW in Friesian Holstein calf was ranged from -3.12 to +2.99 kg. According to Table 3, the highly of RA values (RA>0.80) were reached in the sires with had a large number of progenies. Therefore, low RA value in a bull indicates that selection in this bull is not accurate because of less number of progeny for analysis. (Putra et al., 2015).

CONCLUSION
The h 2 value of BW in Bali cattle at WNT province included of high category and can be described that BW of Bali cattle in this province can be increased based on BW. The r value of BW and CI in Bali cattle at WNT province included of high category and described that cows evaluation can be conducted with two records data of these traits. The cattle with positive of EBV and MPPA values can be used for breeding program in WNT province.