A gene expression estimator of intramuscular fat percentage for use in both cattle and sheep

Table 4 Comparison of the performance of different measures and estimators of IMF%

Method	Number of animals¹	WA control subgroup	WA HGP subgroup	P-value²	Predicted experiment size³
NIRS measured IMF%	141	2.37 ± 1.00⁴	1.90 ± 0.83	0.001	198
Ultrasound estimated IMF%	173	2.66 ± 0.72⁴	2.93 ± 0.54	1.000	N/A⁵
NIRS measured IMF%	22	2.07 ± 0.77⁴	1.79 ± 0.54	0.340	294
IMF% calculated by CIDEA formula	22	2.30 ± 1.19⁶	1.60 ± 1.14	0.080	148
IMF% calculated by IMF 5 gene set formula	22	2.66 ± 1.04⁶	1.30 ± 0.92	0.003	26
Ranking animals using CIDEA	22	9.50 ± 6.62⁷	13.20 ± 6.16	0.100	156
Ranking animals using IMF 5 gene set	22	7.20 ± 5.47⁸	15.10 ± 4.99	0.0026	22
Ranking animals using TAG gene set	22	7.00 ± 4.52⁸	15.30 ± 5.48	0.0017	20
CIDEA DE	22	13.09 ± 0.44⁷	12.83 ± 0.38	0.080	130
IMF 5 gene set DE	22	0.25 ± 0.38⁹	−0.27 ± 0.31	0.0014	24
TAG gene set DE	22	0.28 ± 0.36⁹	−0.27 ± 0.33	0.00076	20

¹Number of animals used for the analysis.
²For the test that average measured or estimated IMF%/gene expression/ranking in HGP-treated animals is lower than in control animals.
³The sample size predicted to be required to observe a significant result (P < 0.05) with 95% confidence intervals.
⁴Mean values, standard deviation and P-values are calculated using original record data.
⁵Ultrasound estimated IMF% did not detect an effect of the expected direction.
⁶Mean values, standard deviation and P-values are calculated using rescaled gene expression values.
⁷Mean values, standard deviation and P-values of ranking and DE of CIDEA are calculated using rescaled gene expression values.
⁸Mean values, standard deviation and P-values of ranking calculation of IMF 5 gene set and TAG gene set are based on Mann-Witney test.
⁹Mean values, standard deviation and P-values of IMF 5 gene set and TAG gene set DE are calculated using rescaled gene expression values.

ISSN: 2049-1891