Skip to main content

Table 1 The publicly available datasets used in the current study

From: A simple method to combine multiple molecular biomarkers for dichotomous diagnostic classification

Characteristic Dataset #1 Dataset #2 Dataset #3 Dataset #4
Authors Petricoin et al, 2002 [19] Gordon et al, 2002 [18] Golub et al, 1999 [17] van't Veer et al 2002 [20]
Dataset alias OvCa LuMe LLML BrCa
Biomarker Proteomic mass spectra Gene expression Gene expression Gene expression
# Biomarkers 15,154 12,533 7,129 24,481
Diagnostic classes Ovarian cancer Normal Lung adeno-carcinoma Mesothelioma Acute lymphocytic leukemia Acute myeloid leukemia Metastasis within 5 years Metastasis after 5 years
N (Total) 162 91 150 31 46 26 34 44
N (Training) 83 49 16 16 27 11 17 26
N (Test set) 79 42 134 15 19 15 17 18