Table 1.

Breast cancer cohorts

GEO Accession No.	Sample Size
GSE3494 [21]	251†
GSE4922 [1]	289†
GSE2990 [22]	189
GSE1456 [23]	159
GSE7390 [24]	198
GSE11121 [25]	200
GSE12093 [26]	136
GSE2603 [27]	121‡
GSE5327 [28]	58
GSE2034 [29]	286

GEO Breast cancer cohorts containing 1887 samples were reduced to generate the 1614 sample dataset. † 248 overlapping between these cohorts were removed to retain unique samples. ‡ 22 cell lines were removed from this cohort keeping patient samples only