Relating past and present diet to phenotypic and transcriptomic variation in the fruit fly

BMC Genomics

Table 1 Analysis of deviance for each phenotype, indicating the relative effect size of adult diet, larval diet and their interaction per sex relative to the null model with no factor effects

Phenotype	Sex	Factor	Log likelihood	χ2	df	p value
Virgin Lifespan	Female	Null model	−22,555.16
		Adult diet (A)	−21,754.99	1600.34	2	<0.001
		Larval diet (L)	−21,679.02	151.94	2	<0.001
		A * L	−21,667.75	22.53	4	<0.001
	Male	Null model	−19,930.17
		Adult diet (A)	−19,757.87	282.18	2	<0.001
		Larval diet (L)	−19,898.96	62.42	2	<0.001
		A*L	−19,747.86	20.01	4	<0.001
Mated Lifespan	Female	Null model	−4927.58
		Adult diet (A)	−4894.87	65.42	2	<0.001
		Larval diet (L)	−4893.09	3.55	2	0.171
		A * L	−4886.69	12.79	4	0.012
	Male	Null model	−3919.84
		Adult diet (A)	−3916.34	3.77	2	0.151
		Larval diet (L)	−3918.23	3.23	2	0.200
		A*L	−3911.46	9.74	4	0.045
Early Fecundity	Female	Adult diet (A)		3100.10	2	<0.001
		Larval diet (L)		94.98	2	<0.001
		A * L		4.49	4	0.343
Late Fecundity	Female	Adult diet (A)		8461.28	2	<0.001
		Larval diet (L)		91.94	2	<0.001
		A * L		53.53	4	<0.001

For Cox proportional hazard models and GLMs with Poisson distribution the chi-squared test is most appropriate

ISSN: 1471-2164