. 2017 Mar 31;7:45412. doi: 10.1038/srep45412

Table 1. Linear mixed models of phenological variables in relation to sex and methylation at Clock poly-Q or 5′-UTR.

	poly-Q					5′-UTR
	χ²₁	F	df	P	Coefficient (SE)	χ²₁	F	df	P	Coefficient (SE)
Departure date from wintering area
Year	0.00					0.00^{^a}
Area	0.00					0.00^{^a}
Sex		10.79	1,62	0.002			0.81^{^a}	1,61	0.371
Methylation		17.43	1,62	<0.001			0.13^{^a}	1,61	0.717
Sex × Methylation		10.20	1,62	0.002			1.54	1,60	0.219
Males					−0.29 (0.24)					0.57 (0.56)
Females					−2.15 (0.53)***					−0.58 (0.75)
Arrival date to breeding site
Year	9.30**					5.40^{^a},^*
Area	3.00					3.70^{^a}
Sex		4.38	1,57	0.041			2.16^{^a}	1,57	0.147
Methylation		17.64	1,57	<0.001			0.60^{^a}	1,57	0.440
Sex × Methylation		3.74	1,57	0.058			0.00	1,56	0.997
Males					−0.59 (0.22)^**					−0.31 (0.52)
Females					−1.61 (0.50)^**					−0.31 (0.67)
Breeding date
Year	7.70^**					7.70^{^a},^**
Area	4.30^*					4.30^{^a},^*
Sex		3.67	1,77	0.059			6.39^{^a}	1,78	0.014
Methylation		4.97	1,77	0.029			2.65^{^a}	1,78	0.108
Sex × Methylation		4.60	1,77	0.035			1.45	1,77	0.232
Males					−0.02 (0.19)					−0.95 (0.47)
Females					−0.97 (0.40)^*					0.14 (0.77)

*0.05 > P > 0.01; **0.01 ≥ P ≥ 0.001; ***P < 0.001. ^aEstimated from a model excluding the interaction term. χ²₁: likelihood ratio test for the random effect of year or area. Methylation (percentage of 5-methyl-cytosine) is included as a covariate. Year and area are included as random factors; sex as a fixed-effect factor. Sample sizes for the analyses of poly-Q were as follows (males, females): departure date from wintering area: 49, 20; arrival date to breeding site: 45, 19; breeding date: 58, 26. The corresponding values for the analyses of 5′-UTR were: 48, 19; 45, 18; 59, 25; 61, 25.