Parameters of synthetic data

Tab. 4, shows the parameters of the procedure selected to generate the artificial datasets used in the experiments. For the meaning of the parameters, see the section Generation of artificial gene expression data.

Table 4: Parameter configurations used to generate the artificial data.

$\mathbf{}$	$\mathbf{Colon}$		$\mathbf{Leukemia}$		$\mathbf{DLBCL-FL}$		$\mathbf{DLBCL-Outcome}$
$\mathbf{}$	$\mathbf{diseased}$	$\mathbf{control}$	$\mathbf{ALL}$	$\mathbf{AML}$	$\mathbf{DLBCL}$	$\mathbf{FL}$	$\mathbf{cured}$	$\mathbf{refractory}$
ES	6	2	6	4	3	2	2	2
$[l_{\min},l_{\max}]$	[5, 7]	[650, 650]	[5, 50]	[5, 1782]	[5, 2000]	[10, 2000]	[5, 1500]	[5, 1500]
d	1	0.01	1	0.01	0.01	1	1	1
s	0.8	0.6	1	1	0.6	0.6	1	0.7
Positive ES	0	0	0	0	0	0	0	0
Negative ES	0	0	0	0	0	0	0	0
Heterogeneous
ES	6	2	6	4	3	2	2	2
Mandatory ES	4	1	6	4	1	1	2	1
n	62		72		77		58
$n_{1}$	40		47		58		32
r	2000		7129		6285		6285
M	4.5
$[x_{min}, x_{max}]$	[-4.5, 4.5]
$[t_{\min},t_{\max}]$	[1.5, 3.5]
Common ES	0
p	0.009
e	0.0005
$\nu$	0.15

Table 5 shows the mean and the standard deviation we used to generate simple gaussian-based distributions of the $5$ series of synthetic datasets.

Table 5: Table containing the mean and the standard deviation of the normal distributions followed by the perturbed genes in the five series of artificial datasets.

	series $1$
	class $1$	class $2$
	vectors $1$ to $40$	vectors $41$ to $62$
Modified components	$1$ to $60$	$1991$ to $2000$
mean	$1$	$-1$
standard deviation	$1$	$1$
	series $2$
	class $1$	class $2$
	vectors $1$ to $40$	vectors $41$ to $62$
Modified components	$1$ to $60$	$1991$ to $2000$
mean	$2$	$-2$
standard deviation	$1$	$1$
	series $3$
	class $1$	class $2$
	vectors $1$ to $40$	vectors $41$ to $62$
Modified components	$1$ to $60$	$1991$ to $2000$
mean	$1$	$-1$
standard deviation	$3$	$3$
	series $4$
	class $1$	class $2$
	vectors $1$ to $40$	vectors $41$ to $62$
Modified components	$1$ to $60$	$1991$ to $2000$
mean	$2$	$-2$
standard deviation	$1$	$1$
	series $5$
	class $1$	class $2$
	vectors $1$ to $40$	vectors $41$ to $62$
Modified components	$1$ to $60$	$1991$ to $2000$
mean	$1$	$-1$
standard deviation	$2$	$2$