Table 2

Descriptive information on datasets extracted from the GEO database used in this study.

Human Independent Samples (Case-control)

GSE Number
GEO Description
Species
Control (N)
Cancer Type (N)
Mean Fold Difference in MAO-A
p-value
Percentage of cancer samples < lowest control

GSE3189
Cutaneous malignant melanoma
H
Normal (7)
Malignant Melanoma (45)
-15.0
4.9*10-21
100%
GSE1037
Lung neuroendocrine tumor classification
H
Normal Lung (19)
Small Cell lung Carcinoma (15)
-9.7
1.1*10-10
100%
GSE2549*
Human Malignant Pleural Mesothelioma
H
Normal Pleura (5)
Malignant pleural mesothelioma (40)
-7.4
4.6*10-10
90%
GSE61
Breast tumor characterization
H
Normal Breast (10)
Basal-like Tumors (10)
-5.6
1.6*10-4
100%
GSE2379
Hypopharyngeal cancer at various stages of progression
H
Normal Uvula (3)
Early Stage (4)
-3.4
7.7*10-4
100%
GSE3744
Basal-like breast cancer tumors
H
Normal Breast (7)
Basal-like Cancer (18)
-5.7
0.002
88%

* Lung tissue and cell lines were excluded. Only primary tumor samples and normal pleura were analyzed.
Paired Samples

GEO Series Number
Cancer Type
Species
Control (N)
Cancer (N)
Fold Difference in MAO-A
p-value
Percentage

GSE3268
Human Squamous Cell Carcinoma of the lung
H
Control (5)
Cancer (5)
-2.7
0.001
100%
GSE781
Clear Cell Carcinoma of the human kidney
H
Normal (7)
Renal clear Cell Carcinoma (7)
-2.2
0.002
100%
GSE2514
Pulmonary adenocarcinoma
H
Adjacent Normal (10)
Tumor (10)
-1.7
0.004
80%
GSE2685**
Gastric cancer
H
Normal (6)
Cancer (6)
-3.6
0.03§
100%
GSE3678
Papillary thyroid cancer
H
Normal (7)
Papillary thyroid Cancer (7)
-1.2
0.08§
71%
GSE3467
Papillary thyroid cancer
H
Normal (9)
Papillary thyroid Carcinoma (9)
-1.1
0.286§
67%

**Paired data was extracted and unpaired samples were excluded.
Animal Models

GEO Series Number
Cancer Type
Species
Control (N)
Cancer (N)
Fold Difference in MAO-A
p-value
Percentage

GSE2514
Urethane-induced lung tumor model of pulmonary adenocarcinoma
M
Adjacent Normal (10)
Tumor (29)
-1.7
6.6*10-10
86%
GSE3519***
Liver cancer model
Z
Normal Liver Tissue (10)
Liver Tumor (10)
-2.2
1.7*10-5
100%
GSE3348
LH overexpressing virgin mice (luteinizing hormone overexpression causes spontaneous mammary tumors)
M
Wild type Breast tissue (3)
LH-overexpressing Breast tissue (3)
-1.7
5.9*10-5
100%
GSE2426
Patched heterozygous model of medulloblastoma
M
Granule cell Precursor (4)
Tumor Cells from Heterozygotes (5)
-1.6
6.9*10-5
100%
GSE422
Colon cancer
M
C57/BL6 wild-type (6)
APC(Min/+) mutant (10)
-18.9
1.1*10-4
100%
GSE1872****
N-methyl-N-nitrosourea-induced breast cancer model (R)
R
Normal (11)
Cancer (9)
-1.7
5.9*10-5
100%
GSE2528
Mammary tumorigenesis in MMTV-neu model
M
Wild type Normal Breast (3)
Mammary tumors in MMTV-neu Model (7)
-1.8
0.001
100%

*** Zebrafish do not have separate MAO-A and MAO-B therefore MAO levels were analyzed.

**** Biological replicates were averaged for the analysis.

Descriptive information on datasets extracted from the GEO database used in this study. Cases are grouped by whether control and cancer tissues came from different human patients (independent samples), the same human patients (paired samples), or animal models. Tissues are from humans (H) mice (Mus musculus, M), rats (Rattus norvegicus, R), or zebrafish (Danio rerio, Z). For each dataset, sample sizes of control and cancer tissues are provided. The fold difference in mean intensities of MAO-A describes the amount of suppression of the expression of that gene in cancerous tissue; for example, a fold difference of -4 indicates 4 times less expression in cancer tissue than in normal tissue. P-values reflect significant fold differences in expression between cancer and control tissues using the appropriate t-test; cancer types are listed in order of decreasing significance of MAO-A expression. All datasets showed a significant change as determined by the sequential Bonferoni-Holm adjustment unless marked § The percentage column contains the percent of individual cancer tissue samples in each data set that had lower levels of MAO-A expression than the lowest single control sample.

Rybaczyk et al. BMC Genomics 2008 9:134   doi:10.1186/1471-2164-9-134