Prediction of lifetime milk production in synthetic crossbred cattle strain Vrindavani of North India

Abstract

Ten-year data records on growth (birth weight-BWt), five initially reproductive traits (age at first calving-AFC, first dry period-FSP, first dry period-FDP, first calving interval-FCI and first lactation length-FLL) along with the part lactation records of 100, 170 and 240days of first lactation (my100-1, my170-1 and my240-1) and second lactation (my100-2, my170-2 and my240-2) and their respective total milk yields (total lactation milk yield of first lactation-TLMY1 and total lactation milk yield of second lactation-TLMY2) were used to predict LTMY5 (lifetime milk yield as total milk yield up to 5 lactations) and LTMY4 (lifetime milk yield as total milk yield up to 4 lactations). It was observed that first calving interval (FCI) happens to be important predictor (out of initially expressed growth-birth weight(BWt), reproductive traits- AFC, FSP, FDP, FCI, FLL and first lactation milk traits- my100-1, my170-1, my240-1 and TLMY1) for lifetime prediction (both LTMY4 and LTMY5). Prediction of LTMY4 and LTMY5 with respect to initial growth (birth weight), reproductive traits (AFC, FSP, FDP, FCI, FLL) and first 2 lactations (my100-1, my170-1 my240-1, TLMY1, my100-2, my170-2 my240-2 and TLMY2) indicated the contribution of my240-2 followed by TLMY1 and my170-2. They jointly explained 40.32% variation in estimated value of LTMY4. However, prediction of LTMY5, with respect to these predictors showed my240-2 together with FLL jointly explained 26.71% variation in estimated value.

Full text

111
India with 199.08 million heads of cattle (33.06 million
crossbreds and 166.02 million indigenous) and 105.34 million
heads of buffaloes (BAHS 2010) is the largest producer of
milk among the countries of the world (112.5 million tones
during 2009–10). The contribution of agriculture to total
gross domestic product has declined (34.72% in 1980–81 to
10.99% in 2008–09) whereas the livestock sector has also
decreased slightly from 4.82% (1980–81) to 3.26% (2008–
09; GOI 2010). However, the contribution of livestock sector
to agricultural GDP has increased from 13.88% (1980–81)
to 29.64% (2009–09, GOI 2010). This indicated towards the
importance of livestock sector in the Indian agriculture. As
per the economic survey for 2011–12, the country’s per capita
milk availability in 2009–10 was at 263 g/day still below the
world average of 279.4 g/day (PTI, February 25, 2011-
ProfitNDTV.com).
Panda et al. (2006), predicted total milk yield based on
most frequent daily milk yield and highest daily milk yield
Present address: 1,2Senior Scientist (tanweer_khan234
@rediffmail.com akstomar2003@yahoo.com), Livestock
Production and Management Section. 3
Joint Direcor
(jdee@ivri.res.in).
Indian Journal of Animal Sciences 82 (11): 1367–1371, November 2012
Prediction of lifetime milk production in synthetic crossbred cattle
strain Vrindavani of North India
T A KHAN1, A K S TOMAR2 and TRIVENI DUTT3
Indian Veterinary Research Institute, Izatnagar, Uttar Pradesh 243 122 India
Received: 10 August 2011 ; Accepted: 11 April 2012
ABSTRACT
Ten-year data records on growth (birth weight-BWt), five initially reproductive traits (age at first calving-AFC, first
dry period-FSP, first dry period-FDP, first calving interval-FCI and first lactation length-FLL) along with the part
lactation records of 100, 170 and 240days of first lactation (my100_1, my170_1 and my240_1) and second lactation
(my100_2, my170_2 and my240_2) and their respective total milk yields (total lactation milk yield of first lactation-
TLMY1 and total lactation milk yield of second lactation-TLMY2) were used to predict LTMY5 (lifetime milk yield as
total milk yield up to 5 lactations) and LTMY4 (lifetime milk yield as total milk yield up to 4 lactations).
It was observed that first calving interval (FCI) happens to be important predictor (out of initially expressed growth
– birth weight(BWt), reproductive traits- AFC, FSP, FDP, FCI, FLL and first lactation milk traits- my100_1, my170_1,
my240_1 and TLMY1) for lifetime prediction (both LTMY4 and LTMY5). Prediction of LTMY4 and LTMY5 with
respect to initial growth (birth weight), reproductive traits (AFC, FSP, FDP, FCI, FLL) and first 2 lactations (my100_1,
my170_1 my240_1, TLMY1, my100_2, my170_2 my240_2 and TLMY2) indicated the contribution of my240_2 followed
by TLMY1 and my170_2. They jointly explained 40.32% variation in estimated value of LTMY4. However, prediction
of LTMY5, with respect to these predictors showed my240_2 together with FLL jointly explained 26.71% variation in
estimated value.
Key words: Vrindavani cattle, Lifetime milk production, Part lactation
(of a month in Sahiwal cows with accuracy. Malhotra and
Singh (1980) predicted lifetime production (total milk yield
in the first 3 lactations) for Red Sindhi cows on the basis of
traits available in early life. Puri and Sharma (1965) studied
first lactation yield and age at first calving on lifetime
production and determined the relative importance of them
for selection purposes in Red Sindhi and crossbred cows.
They have taken yield up to 5 lactations. Prediction of lifetime
milk production using artificial neural network and multiple
regressions was also carried out in Sahiwal cattle (Gandhi et
al. 2000); the traits involved were age at first calving, first
lactation 305-day or less yield, first lactation length, first
service period and first dry period. Shinde et al. (2010)
predicted lifetime milk production up to third lactation in
Phule Triveni cows. Vrindavani cattle, synthetic a crossbred
cattle strain of India developed at IVRI, has exotic inheritance
of Holstein-Friesian, Brown Swiss, Jersey and indigenous
inheritance of Hariana cattle (Singh et al. 2011). Lifetime
production is an important economic parameter when
defining the breeding objectively. Gugger et al. (2007) used
lifetime production (LP, production to sixth lactation) and
productive life (PL, number of completed lactations) to obtain
daughter averages to estimate heritabilities in cattle

1368 KHAN ET AL. [Indian Journal of Animal Sciences 82 (11)
112
populations in Switzerland. The importance raises the
question of whether the performance of a cow in subsequent
lactations is repetitive enough genetically. Performance in
first lactation can contribute useful information also about
later lactations. In this study, it was aimed to find out the
lifetime milk yield prediction models to be used as tools to
evaluate the potential of Vrindavani cattle for selection.
MATERIAL AND METHODS
The present study was undertaken on Vrindavani cattle
maintained at cattle and buffalo farm, Indian Veterinary
Research Institute, Izatnagar, India. Data from 1999 to 2009
(ten-year) on growth (birth weight- BWt), 5 initially
expressed reproductive traits (age at first calving-AFC, first
dry period-FSP, first dry period-FDP, first calving interval-
FCI and first lactation length-FLL) along with the part
lactation records of 100, 170 and 240days of first lactation
(my100_1, my170_1 and my240_1) and second lactation
(my100_2, my170_2 and my240_2) and their respective total
milk yields (total lactation milk yield of first lactation-
TLMY1 and TLMY2) of cattle having more than 99days
FLL, were used in the study. Total milk yield up to 4 lactations
(LTMY4) and up to 5 lactations (LTMY5), were taken as
life time production (Puri and Sharma 1965, Gugger et al.
2007, Shinde et al. 2010).
Step-wise regression analysis for LTMY4 and LTMY5
with 5 initially expressed traits, viz. BWt, AFC, FDP, FSP,
FCI, FLL, and part days milk yield (100, 170 and 240 days)
of first lactation and first 2 lactations (BWt, AFC, FLL, FCI,
FSP, FDP, my100_1, my170_1, my240_1, TLMY1,
my100_2, my170_2, my240_2, TLMY2) as predictors, was
done using Proc Reg procedure of SAS 12.0. In step-wise
procedure, criterion of significance at 5% was fixed.
Prediction equation was evolved on the basis of per cent
variation explained in predicted variable (R2) together with
Cp-statistics and Fit diagnostics of the model Mallows’ (1973;
SAS Institute Inc. 2009).
Nine mathematical models (linear, logarithmic, inverse,
quadratic, cubic, power, compound, S-curve, growth and
exponential) were evolved for lifetime milk yield (LTMY4
and LTMY5) with most important initially expressed growth,
reproduction and milk production traits as indicated by
stepwise regression analysis as predictors based on
significance level (P<0.01) (Draper and Smith 1966 –SPSS
12.0).
RESULTS AND DISCUSSION
Lifetime production (LTMY4)
On the basis of first-lactation data: Step-wise regression
analysis for LTMY4 with respect to BWt, AFC, FLL, FCI,
FSP, FDP, my100_1, my170_1, my240_1, tlmy1, could retain
first calving interval (FCI) as lone predictor (when
significance level was fixed at 5%) and explained 11.63%
variation in estimated value of LTMY4. All variables left in
the model are significant at the 0.1500 level.
LTMY4=10120+ 6.31745** FCI; R2=11.63%; Adj. R2=10.05%
and Cp = 1.9357
*,significant of coefficient (P<0.05) **,significant of
coefficient (P<0.01)
Curve estimation shows appropriateness of quadratic
function in prediction of LTMY4 with FCI as predictor and
could explain 20.30% in estimated values (Fig. 1)
LTMY4=4234.61+24.3589 (FCI) – 0.0137 (FCI)2; R2=20.30%
On the basis of first two lactation records: Stepwise
regression analysis for LTMY4 with respect to BWt, AFC,
FLL, FCI, FSP, FDP, my100_1, my170_1, my240_1,
TLMY1, my100_2, my170_2, my240_2, TLMY2, could
retain part total milk yield at 240 days of second lactation
(my240_2) together with first lactation milk yield (TLMY1)
and part total milk yield at 170 days of second lactation
(my170_2) as predictors and jointly explained 40.32%
variation in estimated value. The my240_2 was the main
contributor (21.625) followed by TLMY1 (12.55%) and
my170_2 (6.15%) respectively.
LTMY4=6702.109+ 0.96292**TLMY1 -3.46840*
my170_2 + 4.23887** my240_2; R2=40.32%; Adj.
R2=36.59%
Fit diagnostics for LTMY4 shows appropriateness of the
model. Examination of residual plots indicated that residuals
appear to be a random scatter around a zero reference line
and display no heteroscedasticity. The quantile plot and
histogram of residuals shows no problem with the normality
assumption. The plot of the observed values verses the
predicted values indicate a good fit for the model. The Fit-
mean residual plot indicates the model accounts for a good
deal of the variability in the LTMY4 (Fig. 2).
Lifetime production(LTMY5)
First lactation data: Step-wise regression analysis for
LTMY5 with respect to BWT, AFC, FLL, FCI, FSP, FDP,
my100_1, my170_1, my240_1, tlmy1, could retain first
calving interval (FCI) as lone predictor (when significance
Fig. 1. Predicted versus Observed Lifetime Milk Yield (LTMY4)
in Vrindavani Cattle.

November 2012] PREDICTION OF LIFETIME MILK PRODUCTION IN SYNTHETIC CROSSBRED CATTLE 1369
113
level was fixed at 5%) and explained 9.96% variation in
estimated value. All variables left in the model are significant
at the 0.1500 level.
LTMY5=12677+ 7.75544* FCI, R2=9.96%; Adj. R2=8.35%
and Cp = -2.3411
Curve estimation shows appropriateness of quadratic
function in prediction of LTMY5 with FCI as predictor and
could explain 16.10% in estimated values (Fig. 3).
LTMY5=7799.26+21.6363 (FCI) – 0.0098 (FCI)2; R2=16.10%
First two-year records: Step- wise regression analysis for
LTMY5 with respect to BWt, AFC, FLL, FCI, FSP, FDP,
my100_1, my170_1, my240_1, TLMY1, my100_2,
my170_2, my240_2, tlmy2, could retain part total milk yield
at 240 days of second lactation (my240_2) together with first
Fig. 2.Fit diagnostics plots for lifetime milk yield: LTMY4.
Fig. 3. Predicted versus observed lifetime time milk yield
(LTMY5) in Vrindavani cattle

1370 KHAN ET AL. [Indian Journal of Animal Sciences 82 (11)
114
lactation length (FLL) as predictors and jointly explained
26.71% variation in estimated value. The my240_2 was the
main contributor (18.74%) followed by FLL (7.97%)
respectively.
LTMY5=6706.13935 + 14.47479* FLL + 2.02141 ** my240_2
R2=26.71%; Adj. R2=23.71%
Fit diagnostics for LTMY5 showed appropriateness of the
model as has been indicated for LTMY4 (Fig. 4).
Puri and Sharma (1965) studied lifetime prediction models
in Tharparkar, Sahiwal, and Red Sindhi breeds and one-half
Jersey × one-half Thari crossbred cows and found good
precision of accuracy based on first lactation yield and age
at first calving. In present study first calving interval (FCI)
found to be important predictor for both LTMY4 and LTMY5
explaining 20.30 and 16.10% in the estimated values, as
quadratic model. This implies that the early expressed traits
had an important role in assessment of lifetime milk yield.
Gandhi et al. (2009) also find appropriateness of age at first
calving (AFC), first lactation 305 day or less milk yield
(FL305DMY), first lactation length (FLL), first service
period (FSP) and first dry period (FDP) and full models
explained 25.92% variation in the estimated values. In the
present study, based on initial growth, reproduction traits
and first two year part-lactation records and respective total
milk yield as predictors, found part total milk yield at 240
days of second lactation (my240_2) together with first
lactation milk yield (TLMY1) and part total milk yield at
170 days of second lactation (my170_2) explained 40.32%
Fig. 4. Fit diagnostics plots for lifetime milk yield: LTMY5.

November 2012] PREDICTION OF LIFETIME MILK PRODUCTION IN SYNTHETIC CROSSBRED CATTLE 1371
115
variation in estimated value of LTMY4. However in LTMY5,
part total milk yield at 240 days of second lactation
(my240_2) together with first lactation length (FLL) jointly
explained 26.71% variation in estimated value.
Based on these findings it is concluded that the models
LTMY4=4234.61+24.3589 (FCI) – 0.0137 (FCI)2,
LTMY4=6702.109+0.96292 TLMY1 -3.46840 my170_2 +
4.23887 my240_2, LTMY5=7799.26+21.6363 (FCI) –
0.0098 (FCI)2 and LTMY5=6706.13935 + 14.47479 FLL +
2.02141 my240_2, can be used as tool for early selection of
crossbred cattle.
ACKNOWLEDGEMENTS
Authors are grateful to Director and Incharge Livestock
Production and Management Section for constant
encouragement and allowing the facilities
REFERENCES
Basic Animal Husbandry Statistics (BAHS). 2010. GOI, Ministry
of Agriculture, Department of Animal Husbandry Dairying &
Fisheries, Krishi Bhavan, New Delhi
Draper N R and Smith H.1966. Applied Regression Analysis. 407
p. Wiley, New York.
Gandhi R S, Raja T V, Ruhil A P and Kumar Amit. 2009. Prediction
of lifetime milk production using artificial neural network in
Sahiwal cattle. Indian Journal of Animal Sciences 79(10): 1038
40.
Gugger, M., Ménétrey, F., Rieder, S. and Schneeberger, M. 2007.
Heritability of lifetime milk yield and productive life and their
relationship with production and type traits in the Simmental,
Swiss Fleckvieh and Red Holstein populations in Switzerland
Paper presented at a Meeting European Federation of Animal
Sciences. Dublin Ireland EAAP 2007 Paper 39.4
www.eaap.org\Previous_Annual Meetings/./S39_04_Gugger.
pdf
Malhotra, P. K. and Singh, R. P. 1980. Estimation of life-time
production in Red Sindhi cattle using ridge-trace criterion.
Indian Journal of Animal Sciences 50(3): 215–18.
Press Trust of India 2011. Economic survey for 2011-12 (February
25, 2011)-ProfitNDTV.com
Puri T R and Sharma K N S. 1965. Prediction of lifetime production
on basis of first lactation yield and age at first calving for
selection of dairy cattle. Journal of Dairy Science 48(4): 462–
67.
Panda K K, Meheta R K and Das B C. 2006. Prediction of total
lactation milk yield based on most frequent daily milk yield
and highest daily milk yield of a month in Sahiwal cows. Indian
Journal of Animal Sciences 76 (10): 851–52.
Singh R R, Dutt Triveni, Kumar Amit, Tomar A K S and Singh
Mukesh. 2011. On-farm characterization of Vrindavani cattle
in India. Indian Journal of Animal Sciences 81(3): 267–71.
Shinde N V, Mote M G, Khutal B B and Jagtap D Z. 2010. Prediction
of lifetime milk production on the basis of lactation traits in
Phule Triveni crossbred cattle. Indian Journal of Animal
Sciences 80(10): 968–88.
SPSS Inc. 2003 User’s Guide. SPSS Inc., Chicago
SAS Institute Inc. 2009. SAS/STAT ® 9.2User ’s Guide, Second
Edition. Cary, NC: SAS Institute Inc.