Table 1:

Overview of collected datasets

Dataset	GEO Accession	No. tumor cells	No. non-tumor cells	No. patients	Protocol
Tirosh et al. [22]	GSE70630	4,045	302	6	Smart-seq2
Venteicher et al. [23]	GSE89567	5,284	1,057	10	Smart-seq2
Darmanis et al. [24]	GSE84465	1,091	2,498	4	Smart-seq2
Yuan et al. [17]	GSE103224	25,056	4,194	10	Proprietary microwell platform
Neftel et al. [25]	GSE131928	6,863	1,067	28	Smart-seq2
Yu et al. [26]	GSE117891	3,211	2,392	13	STRT-seq
Total		45,550	11,510	71

STRT-seq: single-cell tagged reverse transcription sequencing.