Drug candidate identification based on gene expression of treated cells using tensor decomposition-based unsupervised feature extraction for large-scale data

Table 2 Compound–gene interactions presented in Table 1 that significantly overlap with interactions described in two datasets

For each compound in the table, the upper row: the drug2gene.com dataset was used for comparisons [69], the lower row: the DSigDB dataset was used for comparisons [70]. Columns represent cell lines used in the analysis: (1) BT20, (2) HS578T, (3) MCF10A, (4) MCF7, (5) MDAMB231, (6) SKBR3, (7) A549, (8) HCC515, (9) HA1E, (10) HEPG2, (11) HT29, (12) PC3, (13) A375. ○: a significant overlap between the datasets (P<0.05); ×: no significant overlap between the datasets; —: no data; blank: no significant dose–response relation was identified. The confusion matrix and a full list of commonly selected genes are available in Additional file 3

ISSN: 1471-2105