Table 1.

Empirical examples of top-1 search results. Four-character alphanumeric codes are patient identifiers of the TCGA barcode that is given to each sample.

Dataset	Query	Top-1	Similar clinical features
OV	1331	2548	Same clinical stage (Stage IIIC) and histologic grade (G3).
LUAD	4244	7724	Both reformed smoker for less or equal to 15 years
UCEC	A0GQ	A18A	Same histological type (endometrioid endometrial adenocarcinoma); close clinical stages (ib and ia).
GBM	0003	5411	Same histological type (untreated primary (de novo)).
BRCA	A0SF	A0CZ	Both positive estrogen receptor status and positive progesterone receptor status; same historical type (Infiltrating Ductal Carcinoma).