Table 3.

Sparse Partial Least Squares (sPLS) regression model applied on the gene expression values (Fig. 2) to elucidate their explanatory power resolving in fruit yield values under glasshouse and polytunnel conditions on full-expanded leaves and mature red fruits.

		sPLS's variable importance in projection (VIP)—coefficients
		Glasshouse (Experiment 1)									Polytunnel (Experiment 2)
	Data matrix:	Leaf and fruit			Leaf			Fruit			Leaf and fruit			Leaf			Fruit
	C.D	0.713			0.564			0.617			0.802			0.644			0.541
Tissue	Gene
Leaf	mMDH	0			0						0			1.32
	SBP	0			0						0			0
	SPA	3.50			3.13						1.88			1.78
	PP	0			0						2.45			1.74
	GS2	0			0						0			0
	GLDH	0			0						0			0
	Sweet11										0			0.20
	SUC2	0			0						0			0
	INVINH	0			0						2.40			2.21
	CAT9	0			0						0			0
	AAP1	0.53			0.48						0			0
Fruit	mMDH	0						0.93			1.29						1.38
	SPA	1.72						1.93			1.22						0.95
	GLDH	0						0.32			0.99						1.22
	SUC2	1.89						2.03			0.56						1.49
	STP6	0						0			0						0
	STP3	0						0			0						0.30
	LIN5	0						0			1.70						1.61
	INVINH	0						0			0						0.60
	SUS1	0						0			0						0.41
	AgpL1	0						0			0						0.63
	TMT1	0						0			0						0.61
	AAP6	2.23						2.22			2.52						2.06
	CAT9	0						0.49			0						1.24
	SUC9	0						0			0.11						0.73

Values represent sPLS's Variable Importance in Projection (VIP)—coefficients. Threshold for significative value has been arbitrary fixed in 1.2 and coefficients above this limit are set in bold face.