Differentiation between Acute Skin Rejection in Allotransplantation and T-Cell Mediated Skin Inflammation Based on Gene Expression Analysis

Table 2

Results from two-way analysis of variances (ANOVA) of log2-fold changes of gene expression considering the skin type (hairy or nonhairy skin) and the model (allograft rejection or inflammatory skin disease) as factors and their potential interaction.

Gene	Factor	Df	Sum Sq	Mean Sq	value		Sign.
Il12b	SKIN_TYPE	1	28.20	28.20	35.38	2.0 × 10^−5	***
	MODEL	1	25.60	25.60	32.13	3.5 × 10^−5	***
	SKIN_TYPE:MODEL	1	0.26	0.26	0.33	0.57
Il17a	SKIN_TYPE	1	211.69	211.69	65.61	4.7 × 10^−7	***
	MODEL	1	54.56	54.56	16.91	8.2 × 10^−5	***
	SKIN_TYPE:MODEL	1	17.54	17.54	5.44	0.033	*
Il1b	SKIN_TYPE	1	5.95	5.95	4.32	0.054	.
	MODEL	1	19.71	19.71	14.31	0.0016	**
	SKIN_TYPE:MODEL	1	14.00	14.00	10.17	0.0057	**
Il12a	SKIN_TYPE:MODEL	1	6.76	6.76	18.63	5.3 × 10^−4	***
Il2	SKIN_TYPE:MODEL	1	16.18	16.18	17.04	7.9 × 10^−4	***
Tnf	SKIN_TYPE:MODEL	1	11.82	11.82	22.81	2.1 × 10^−4	***

Sign: significance score: (, , , and ), Df: degree of freedom.
Sum Sq: sum of squares, Mean Sq: mean of sum of squares.
Log2-fold changes were related to the mean of the controls in the respective groups. Results are shown for genes with most significant effect of the factor model (rejection versus inflammation) and only moderate interaction with the factor skin type, as well as for genes with the most significant statistical interaction between the two factors.