Meta-analysis of studies related to genetic parameters for productive and reproductive traits in Iranian native fowl

Introduction: Native fowl have become valuable genetic resources over the years due to their adaptation to harsh conditions in rural areas (Emamgholi et al 2009). Native fowl, despite their lower growth rate and lower egg production compared with broiler strains such as Ross and Cobb and layer strains such as Hy-Line and Lohman, perform much better in the rural system due to their good disease resistance and production under adverse nutritional and environmental conditions (Lakhi et al 2013). Breeding programs have a significant effect on improving the genetic structure of native fowl (Kiani manesh et al 2002). The purpose of this study was to combine the results of studies related to estimating genetic parameters for productive and reproductive traits in native fowl using meta-analysis method in order to improve the accuracy of estimations. In addition, the use of meta-analysis method eliminates the existing differences between studies by examining the heterogeneity between them (Field and Gillett 2010; Lortie et al 2013; Hooijmans et al 2014). Increasing the accuracy of estimates can consequently increase genetic improvement through selecting proper candidates to be parents of next generations. Higher performance will increase the courage of farmers to keep native fowl and therefore it help to keep the birds as a valuable genetic source for the future.
Materials and Methods: In this meta-analysis study, previous studies in the field of estimating genetic parameters of productive and reproductive traits in Iranian native fowl were collected and used. These articles consisted of 33 articles which were published between years 2007 and 2016. Information used in current study included heritability, genetic and phenotypic correlations of productive and reproductive traits. These traits consisted of birth weight, weight at eight weeks of age, weight at twelve weeks of age, egg number, egg weight, age at sexual maturity and weight at sexual maturity. Cochran test and I2 criterion were used for testing heterogenty among used articles. The range of heterogeneity among studies could be from zero to 100 percent. Ranges of 0 to 25 percent, 25 to 50 percent, 50 to 75 percent and 75 to 100 percent indicate very low, low, moderate and high heterogeneity among studies, respectively. Fixed meta-analysis model is usually recommended for very low and low heterogeneity among studies and random meta-analysis model is normally used for moderate to high rates of heterogeneity among studies. Due to high heterogenty observed among studies in present study, random effects model meta-analysis was performed using Metacor package R software version 3,3,1 and Comprehensive Meta-Analysis (CMA) software version 3. The weighted average of the heritability, genetic and phenotypic correlations were calculated accordingly.
Results and Discussion: According to the results, the highest heritability among productive and reproductive traits was related to egg weight (0.42) and then weight at sexual maturity (0.41). The lowest heritability was related to egg number with a value of 0.20 and then birth weight with a value of 0.25. The highest positive genetic correlation between productive and reproductive traits was estimated between weight at eight and weight at twelve weeks of age (0.86) and then between weight at twelve weeks of age and weight at sexual maturity (0.64). The highest negative genetic correlation was related to the correlation between age at sexual maturity and egg number (-0.66), then egg number and egg weight (-0.19). Due to combing the results of individual studies using meta-analysis, the standard error of estimations for genetic parameters and consequently their 95% confidence interval were significantly reduced compared to individual studies (Bayssa et al 2021). This reduction indicates an increase in the accuracy of the estimates due to aggregating the results of individual studies (Field and Gillett 2010). A comparison between the results of independent studies and the results of meta-analysis shows a significant reduction in estimated standard errors for heritability and genetic and phenotype correlations in the meta-analysis method. For example, the amount of estimated standard error related to heritability of weight at twelve weeks of age in the meta-analysis was calculated to be 0.0009, while the range of reported standard error in the articles for the heritability of this trait was from 0.05 to 0.08. For weight at eight weeks of age the amount of standard error of heritability in the meta-analysis was estimated to be 0.0009, but the range reported for standard error of heritability for this trait in the articles was zero to 0.06. Similar reduction in the amount of estimated standard error were observed for reproductive traits. For example for egg number, the amount of error in the meta-analysis was estimated to be 0.0013, but the range reported in the articles was 0.01 to 0.13. The amount of estimated standard error related to egg weight in the meta-analysis was calculated to be 0.0009, while the range reported in the studied articles was from 0.004 to 0.07. This decreasing trend was similarly observed for other studied traits, which is due to the aggregation of the results obtained from independent studies.

Conclusion: The results of this study showed that the use of meta-analysis method by aggregating the results of individual studies, will increase the accuracy of genetic parameters of productive and reproductive traits in native fowls through increasing the sample size and reducing estimated standard errors. Therefore, by taking into consideration that Iranian native fowl are economically important animals and considerable demands for their products in the country, using genetic parameters obtained from this study with higher accuracy can play an effective role in the successful design of breeding programs for native fowls and promote genetic improvement in native fowl in Iran.