Skip to main content

Table 1 TCGA binary tasks

From: Standard machine learning approaches outperform deep representation learning on phenotype prediction from transcriptomics data

ProjectDiseaseLabelLabel typeGroupSamples
TCGA stage tasks
COADcolon adenocarcinomaII- vs. III+binarytrain505
KIRCkidney renal clear cell carcinomaII- vs. III+binarytrain544
LIHCliver hepatocellular carcinomaI- vs. II+binarytrain374
LUADlung adenocarcinomaI- vs. II+binarytrain542
SKCMskin cutaneous melanomaII- vs. III+binarytrain249
STADstomach adenocarcinomaII- vs. III+binarytrain416
THCAthyroid cancerI- vs. II+binarytrain513
UCECuterine corpus endometrial carcinomaI- vs. II+binarytrain554
LUSClung squamous cell carcinomaI- vs. II+binaryvalidate504
BRCAbreast invasive carcinomaII- vs. III+binarytest1134
TCGA grade tasks
CESCcervical squamous cell carcinomaII- vs. III+binarytrain306
KIRCkidney renal clear cell carcinomaII- vs. III+binarytrain544
LGGlow grade gliomaII- vs. III+binarytrain532
LIHCliver hepatocellular carcinomaII- vs. III+binarytrain374
PAADpancreatic adenocarcinomaII- vs. III+binarytrain179
STADstomach adenocarcinomaII- vs. III+binarytrain416
UCECuterine corpus endometrial carcinomaII- vs. III+binarytrain554
HNSChead-neck squamous cell carcinomaII- vs. III+binarytest504
  1. The 18 binary tasks derived from TCGA used to train supervised models and validate the unsupervised embeddings. The tasks are grouped into two categories, TCGA tumor stage tasks (10), and TCGA tumor grade tasks (8). The project names correspond to those in Fig. 2