Table 3.

Results of the negative binomial logistic regression for the QTLs found by QTL mapping and association analysis. Different statistical models using trichome density (TD) after 6,7 or 8 weeks after transplanting and either a single predictor or an additive model are shown. Independent variables used correspond to the number of alleles of the high density parental at the marker with the highest statistical support. For each model, χ² goodness of fit p-value of the model (GOF), Nagelkerke R², explained deviance (D²) and point estimated of incidence rate ratio (IRR) and 95% confidence interval are reported. *** p-value < 0.001.

QTL Mapping
Phenotype	Model	GOF	R2	D2	Variable	IRR (95% CI)	Sig
TD 6th w	TIVd9	0.361	0.18	0.16	SL2.50ch09_2542786	2.61 (1.81–3.78)	***
	TIVd11	0.348	0.08	0.07	SL2.50ch11_56297460	0.54 (0.37–0.78)	***
	TIVd9 + TIVd11	0.386	0.30	0.26	SL2.50ch09_2542786	2.79 (1.96–4.03)	***
					SL2.50ch11_56297460	0.47 (0.33–0.67)	***
TD 7th w	TIVd9	0.145	0.26	0.21	SL2.50ch09_3756370	3.24 (2.24–4.72)	***
	TIVd11	0.144	0.09	0.08	SL2.50ch11_56297460	0.49 (0.33–0.73)	***
	TIVd9 + TIVd11	0.149	0.37	0.29	SL2.50ch09_3756370	3.24 (2.28–4.63)	***
					SL2.50ch11_56297460	0.48 (0.34–0.69)	***
TD 8th w	TIVd9	0.084	0.21	0.17	SL2.50ch09_2309105	2.88 (1.98–4.17)	***
	TIVd11	0.076	0.07	0.06	SL2.50ch11_56297460	0.56 (0.38–0.82)	***
	TIVd9 + TIVd11	0.091	0.33	0.26	SL2.50ch09_2309105	3.22 (2.24–4.65)	***
					SL2.50ch11_56297460	0.48 (0.33–0.68)	***
Association
Phenotype	Model	GOF	R2	D2	Variable	IRR (95% CI)	Sig
TD 6th w	TIVd9	0.361	0.18	0.16	SL2.50ch09_2542786	2.61 (1.81–3.78)	***
TD 7th w	TIVd9	0.147	0.25	0.20	SL2.50ch09_3727914	3.19 (2.21–4.64)	***
TD 8th w	TIVd6	0.090	0.22	0.17	SL2.50ch06_154414	2.47 (1.79–3.41)	***
	TIVd9	0.085	0.22	0.18	SL2.50ch09_2492691	2.94 (2.03–4.26)	***
	TIVd6 + TIVd9	0.085	0.32	0.25	SL2.50ch06_154414	1.86 (1.33–2.60)	***
					SL2.50ch09_2492691	2.16 (1.44–3.22)	***