Skip to main content

Table 5 Predictive ability accuracy for cross validation scenarios: masking across or within family and temporal masking

From: A promising resilience parameter for breeding: the use of weight and feed trajectories in growing pigs

Trait

Across Family

Within Family

Temporal

BLUP

ssGBLUP

BLUP

ssGBLUP

BLUP

ssGBLUP

ADG

0.22

0.42

0.54

0.62

0.27

0.42

AFI

0.22

0.52

0.45

0.64

0.26

0.57

FCR

0.38

0.56

0.61

0.69

0.31

0.50

A

0.12

0.19

0.23

0.27

0.12

0.27

B

0.13

0.30

0.37

0.47

0.17

0.44

K

0.22

0.37

0.50

0.56

0.19

0.41

Fat

0.36

0.62

0.55

0.69

0.21

0.45

Muscle

0.31

0.58

0.51

0.66

0.31

0.55

lnvarweight

0.27

0.36

0.60

0.57

0.39

0.42

lnMSEweight

0.27

0.34

0.60

0.57

0.40

0.44

lnvarweight_standardized

0.20

0.34

0.49

0.52

0.26

0.34

Skewweight

0.00

0.12

0.23

0.23

0.12

0.23

Lag1weight

0.20

0.28

0.40

0.44

0.12

0.28

Straightness

0.25

0.38

0.58

0.58

0.36

0.48

Mean speed

0.29

0.40

0.53

0.60

0.31

0.49

lnMSEFI

0.33

0.52

0.62

0.68

0.41

0.58

lnMSEdur

0.39

0.60

0.56

0.68

0.53

0.71

lnMSEn_visit

0.37

0.53

0.59

0.64

0.42

0.53

QRFI

0.26

0.36

0.46

0.52

0.46

0.49

QRdur

0.32

0.47

0.52

0.60

0.45

0.60

  1. Predictive ability accuracy was estimated by dividing the predictive ability correlation by the square root of the estimated heritability. These predictive abilities as a correlation and standard deviation of these estimates are provided in Additional file 4: Table S2. For temporal masking, there was only one estimate, and hence, no standard deviation was calculated. BLUP: Best linear unbiased prediction. Genetic parameters estimated with pedigree relationships; ssGBLUP: single-step genomic BLUP: genetic parameters estimated with single-step genomic evaluation. ADG: average daily gain; AFI: average feed intake; FCR: feed conversion ratio; A, B and k: Gompertz growth curve parameters; lnvarweight: natural logarithm of variance of observed versus predicted weights; lnMSEweight: natural logarithm of mean squared error of weight in function of age; lnvarweight_standardized: natural logarithm of variance of standardized weights; skewweight: skewness of observed versus predicted weight distribution; lag1weight: lag1 autocorrelation of observed versus predicted weight distribution; straightness: straightness index of weight in function of age after trajectory analysis; mean speed: mean speed of weight in function of age after trajectory analysis; lnMSEFI: natural logarithm of mean squared error of feed intake in function of age; lnMSEdur: natural logarithm of mean squared error of visit duration in function of age; lnMSEn_visit: natural logarithm of mean squared error of number of daily visits in function of age; QRFI: number of days with feed intake below 5% of quantile after quantile regression; QRdur: number of days with visit duration below 5% of quantile after quantile regression