| 摘要: | 母豬生產力綜合母豬的分娩與哺育能力，直接反應豬場的生產與經濟效益之概況。台灣大多牧場系譜與資料不全，停留在傳統的表型選拔造成繁殖性能無法獲得顯著的改善。本篇研究透過有效完整地蒐集繁殖資料，討論遺傳、環境因素對繁殖性能的影響，分析繁殖性狀的遺傳介量，選擇適當性狀作為母豬生產力的評估指標。試驗於台灣一個以美國品系建立的種豬場進行，收集自2018年5月至2022年12月母豬群的系譜、配種及分娩資料，包括1,149胎藍瑞斯、1,096胎約克夏及252胎杜洛克。分析的繁殖性狀包括：出生總仔數(litter size at birth, LSB)、出生活仔數(litter size at born alive, LSA)、5日齡活仔數(litter size at d 5, LS5)、7日齡活仔數(litter size at d 7, LS7)、死胎數(stillbirth, STILL)、木乃伊數(mummy, MUMMY)、出生窩重(litter weight at birth, LBW)、21日齡窩重(litter weight at d 21, L21DW)、平均出生重(average body weight at birth, ABW)、平均21日齡重(average body weight at d 21, A21DW)及離乳前死亡率(pre-weaning mortality rate, PWM)。所有資料採用混合模式(Mixed Model)，以BLUPF90育種軟體探討品種、胎次及配種季節對繁殖性狀的影響；與分析繁殖性狀的變方與共變方成分，以估計遺傳率、遺傳與表型相關。品種對窩仔數與窩重性狀有影響(P < 0.05)，藍瑞斯表現最好。胎次對LSB、LSA及窩重性狀有影響(P < 0.05)；各品種繁殖性狀表現第二至五胎為佳。11 ~ 4月配種有較多的LSB且ABW與LBW較重(P < 0.05)；5 ~ 10月配種的A21DW與L21DW較重(P < 0.1)。窩仔數性狀遺傳率為0.12 ~ 0.17。其中LS7的遺傳率，藍瑞斯為0.17、約克夏為0.11及杜洛克為0.15。窩重性狀遺傳率為0.14 ~ 0.47，僅L21DW較低為0.14。窩仔數性狀間的相關為0.85 ~ 0.99，與LBW、L21DW的相關為0.50 ~ 0.77。應用LS7與各性狀的相關，以LS7作為選拔指標，可改進LS21、減少藍瑞斯PWM、降低約克夏STILL；並結合L21DW提升離乳窩重。本研究提供台灣三個品種母豬繁殖性狀相關訊息，建立台灣繁殖性能的關鍵指標，進一步實證應用此指標遺傳改進的效率；同時由於品種間有特異性，各品種豬隻繁殖性能評估與選拔方法需予考量。
Sow productivity combines the sow's farrowing and nursing abilities, which directly reflects the overall productivity and economic benefits of a pig farm. Due to not having complete pedigrees and performance information, phenotypic selection is still used in pig populations in Taiwan, resulting in reproductive performance failed to improve significantly. The purpose of this study was to collect complete reproductive data, investigate the impact of different genetic and environmental factors on reproductive traits, analyze the genetic parameters of reproductive traits, and select appropriate traits as a key performance indicator for sow productivity. The study collected pedigree, mating and farrowing data from a Taiwanese breeding farm from May 2018 to December 2022, including 1,149 Landrace, 1,096 Yorkshire and 252 Duroc sows. The fundamental population of this farm is from USA. Reproductive traits including litter size at birth (LSB), litter size at born alive (LSA), litter size at d 5 (LS5), litter size at d 7 (LS7), stillbirth (STILL), mummy (MUMMY), litter weight at birth (LBW), litter weight at d 21 (L21DW), average body weight at birth (ABW), average body weight at d 21 (A21DW), and pre-weaning mortality rate (PWM) were analyzed. All data were analyzed using Mixed Model via BLUPF90 breeding software to study the effects of breed, parity, and mating season on reproductive traits; and analyze the variance and covariance components of reproductive traits, to estimate the heritability as well as, genetic and phenotypic correlation. Breed effect was detected on litter size and litter weight traits (P < 0.05), with Landrace performed the highest. Parity effect had an impact on LSB, LSA and litter weight traits (P < 0.05). The reproductive traits of the three breeds were performed better in the second to fifth parity. There were more LSB and heavier ABW and LBW when mating from November to April (P < 0.05); and were heavier on A21DW and L21DW when mating from May to October (P < 0.1). Estimates of heritability ranged from 0.12 ~ 0.17 for litter size traits. Heritability of LS7 is 0.17 for Landrace, 0.11 for Yorkshire and 0.15 for Duroc. Heritability ranged from 0.14 ~ 0.47 for litter weight traits, only L21DW is lower at 0.14. The genetic correlation between litter size traits was 0.85 ~ 0.99, and the genetic correlation with LBW and L21DW was 0.50 ~ 0.77. Study the relationship between LS7 and various traits, using LS7 as a selection indicator can improve LS21, reduce Landrace PWM, and reduce Yorkshire STILL; and further combined with L21DW to increase litter weight at weaning. The study provides genetic information on reproductive traits for the three breeds and develop a key performance indicator to improve sow productivity in Taiwan. Further research requires to realize the genetic improvement using the indicator. In addition, the genetic parameter estimates showed the breed specificities in reproductivity traits, methods of genetic evaluation and selection need to be further considered. |
