Phenotypic Characterization and Weight Estimation of Young Thalli Sheep Through Multivariate Traits

The aim of this study was to evaluate the multivariate traits in Thalli sheep for phenotypic characterization and their association with the live bodyweight of the animals for the selection of superior individuals by using the principal component analysis (PCA) technique and regression equation. Twenty-one (21) multivariate traits were taken of young Thalli sheep aged 0-12 months in Thall area of Punjab, Pakistan. Morphometric traits measured were withers height (WH), head width (HW), head length (HL), body length (BL), ear width (EW), ear length (EL), neck width (NW), neck length (NL), heart girth (HG), rump length (RL), tail length (TL), rump width (RW), barrel depth (BD), birth weight (BiW), live body weight (BW), sacral pelvic width (SPW), teat diameter (TED), teat length (TEL), testes length (TsL), scrotal circumference (SC) and testes width (TsW). Positive and highly significant (P≤0.01) correlations among different morphometric traits were present in overall animals of 0-12-month age groups of Thalli sheep such as WH, BL, HL, HW, EL, EW, NL, NW, HG, RL, BD, SPW, BiW and BW. The results of principal component analysis of data for all age groups of young Thalli sheep were showed that three principal components (PCs) were extracted as PC1, PC2, and PC3. For the 0-3-month age group, PC1 showed high variance before and after rotation (44.33% and 37.68%), for 4-6-month age group, PC1 had a high value of variance as 58.45% and PC2 had 15.17%, for 7-9-month age group, three principal components were extracted with high eigenvalues (7.56) for PC1 and high variance 50.45% and for 10-12-month age group, PC1 had a high value of variance as 50.05% and PC2 and PC3 had values as 10.67% and 9.05% respectively and their cumulative variance was 69.79%. Regression equations for all age groups of young Thalli sheep presented a very close relationship between predicted bodyweight and actual live bodyweight. The present study showed that hearth girth, body length, and withers height are very important parameters for the estimation of body weight and phenotypical characterization.


F i r s t A r t i c l e
breeding programs in meat animals (Younas et al., 2013). In developing countries, record-keeping is at the initial level and the records about pedigree and progeny of individuals are insufficient and do not provide the estimation about genetic parameters. Therefore, phenotypic characterization is compulsory for the explanation of relationships among linear type traits (Ali et al., 1995). Bodyweight is necessary to make decisions related to feeding, health care, and selection (Slippers et al., 2000).. An interdependence technique, principal component analysis (PCA) has been used by different scientists to explain the sets of variables in the analysis of data that are highly interrelated (Yunusa et al., 2013). The principal component analysis technique is used to identify the body size, body shape, head size, and overall phenotypic characterization in Zulu sheep (Mavule et al., 2013). Phenotypic characterization can be based on animal conformation and genetic parameters can be evaluated by using morphometric traits because these traits play vital role in selection of superior animals (Akbar et al., 2022). Evaluation of multivariate traits in Thalli sheep and their association with the live bodyweight of the animals for the selection of elite individuals is the main objective of the present study using principal component analysis technique and regression equation.

Experimental site
The data were collected from two livestock farms; Livestock Experiment Station, Rakh Ghulaman, District Bhakkar, and Small Ruminant Research and Development Centre, Rakh Khairewala, District Layyah, Punjab, Pakistan. The temperature of both areas is raised during the summer season, which may reach up to 50-55°C during day time. Natural grazing procedure has been adopted during day time and concentrate ration has also been offered to the animals as supplementary feed.

Data collection
The data of Thalli sheep were recorded on the basis of multivariate traits. One hundred and sixty-three young Thalli sheep were selected for data collection. A measuring tape was used to record the different multivariate traits and digital weighing balance was used for the determination of live body weight (Iqbal et al., 2014). Standards for phenotypic characterization of breeds as described by FAO (2012), the following parameters were taken.

Data structure
Animals of different age groups were reared at both research stations. Animals were divided into different groups according to their age as 0-3, 4-6, 7-9 and 10-12 (Iqbal, 2010). Each group was further divided into two sub-groups of male and female animals (Table I).

Statistical analysis
The normality of overall data was checked through a normality test using SPSS 20.0 and all animals were fallen in ±3SD. SPSS software 20.0 was used for statistical analysis of morphometric traits. Data were analyzed statistically for mean, coefficient of variation, standard deviation, and Pearson's coefficient of correlation among different multivariate traits was estimated and data were generated for principal component analysis (PCA) from the correlation matrix. Regression equations were used for the estimation of live body weight.

Descriptive statistics of young Thalli sheep
Descriptive statistics (mean, standard deviation, and coefficient of variation) of multivariate traits of all young Thalli sheep are given in Table II. Coefficient of variation (CV) for 0-3-month lambs of linear body measurements ranged 10-15% but its percentage was 18.52, 21.24, and 22.20 for TL, BiW, and BW as the range of these traits were 8.89cm 2kg and 11kg, respectively. For animals of 4-6-month age group, CV of linear body measurements was ranged from 10-20% and BW, TL, and RW had medium values as 25.13%, 27.40%, and 28.76% respectively. For animals of 7-9-month age group, CV of different multivariate traits ranged from 8.18-21.51% (BW). For animals of 10-12-month age group, CV of for linear body measurements was ranged 10-20% and HW, BW, TL, and RW had medium values as 21.54% 24.54%, 21.09%, and 24.02% as their ranges were 6.6cm, 1kg, 8.89cm, and 13.21cm, respectively. Yakubu et al. (2011) have reported WH and BL as 63com (7-9 age group) and 57cm (4-6 age group), respectively. Yakubu et al. (2011) has reported CV of RW and WH as 13% (0-3 age group) and 9.7% (10-12 age group), respectively. The present study is also similar to that of Lopez-Carlos et al. (2010) in Blackbelly lambs. The results of current are not in line with Iqbal et al. (2014) as means of HG, BL and WH as 47.49cm, 44.60cm, and 49.10cm, respectively. Younas et al. (2013) reported the means of BW, HG, and WH 10kg,48.17cm and 48.85cm respectively, these variations are indication of age and breed differences.

Correlation coefficients among multivariate traits of young Thalli sheep
Correlation coefficients of multivariate traits of 0-3, 4-6, 7-9, and 10-12 month age groups of Thalli sheep are shown in Supplementary Tables I-IV Likewise, correlation coefficients of morphometric traits of 7-9-month age group animals of Thalli sheep had highly significant (P≤0.01) and positive correlations among WH, BL, HL, HW, EL, EW, NL, NW, HG, RL, SPW, RW, BD and BW (0.47-0.86). The findings of the present study are similar to reported by Salako (2006) and results are not in line as reported by Younas et al. (2013) which may be due to breed differences.

Principal component analysis of multivariate traits of young Thalli sheep
Principal component analysis of 0-3, 4-6, 7-9, and 10-12 month age groups of Thalli sheep are shown in Table  VII The results of 0-3 month age group showed that four principal components (PC) were extracted with eigenvalues greater than 1 for principal component 1 (PC1), principal component 2 (PC2), and principal component 3 (PC3). PC1 showed a high variance before and after rotation (44.33% and 37.68%). Yakubu et al. (2011) reported similar results as four principal components were extracted with high values of commonalities. The results of present study are not in line with those of Salako (2006) which may be due to the difference in the environmental area.
For 4-6 month age group animals, there were two PC extracted and PC1 had a high value of variance as 58.45% and PC2 had 15.17% and their cumulative variance was 73.63%. Yakubu et al. (2011) supported the results of current study. Lopez-Carlos et al. (2010) reported low values of morphometric traits in PC1 which are not in line with current study. These differences may occur due to environmental factors. For 7-9 month age group animals, three PCs were extracted with high eigenvalues (7.56) for PC1 and high variance of 50.45%. PC2 and PC3 had variances of 14.18% and 7.86%, respectively and their cumulative variance was 72.50%. Khan et al. (2014) reported that BL and WH had major contributions in PC which supports the results of the present study.
For animals of 10-12 month age group, PC 1 had a high value of variance as 50.05% and PC2 and PC3 had values as 10.67% and 9.05% respectively and their cumulative variance was 69.79%. WH, HG, and BL had high variance and high commonalities, and our results are also supported by Mavule et al. (2013), Khan et al. (2014), Eyduran et al. (2013) and Birteeb and Lomo (2015).

Regression equation for young Thalli sheep
Regression equations for 0-3, 4-6, 7-9, and 10-12 month age groups of Thalli sheep were studied as shown in Table I. The predicted weight and actual BW of all age groups of animals were very close which showed the efficiency of results. The present study showed that HG, BL, and WH are very important for the estimation of BW.
A strong relationship has been determined between BL and BW (Tadesse and Gebremariam, 2010;Yakubu and Mohammed, 2012), between HG and live BW in sheep (Birteeb and Ozoje, 2012;Mavule et al., 2013;Mohammed and Amine, 1996;Suhaila et al., 2013). HG explains a close estimation of BW without using a weighing scale.

CONCLUSION
From the current study, it was concluded that multivariate traits such as HG, WH, and BL had high correlations with BW and among each other in almost all age groups of Thalli sheep. PCA of multivariate traits was explained the variations by PC1 and in some groups, PC2 and PC3 had also some effects. PC1 for WH, BL, and HG had high values which showed maximum variations. Multivariate traits are used for the selection of genetically superior animals for breed improvement and estimation of live body weight in field conditions, so that young animals could be selected early in lives to show their maximum performance.

ACKNOWLEDGEMENT
Special thanks to the management of Small Ruminant Research and Development Centre, Rakh Khairewala, District Layyah and Livestock Experiment Station, Rakh Ghulaman, District Bhakkar, Punjab, Pakistan.

Supplementary material
There is supplementary material associated with this article. Access the material online at: https://dx.doi. org/10.17582/journal.pjz/20200608090631

Statement of conflict of interest
The authors have declared no conflict of interest.

O n l i n e F i r s t A r t i c l e
Supplementary

O n l i n e F i r s t A r t i c l e
Supplementary