Table 1. Conversion of Different Data Types to High and Low States.

Data Type	Attribute	Transformation	High State in Sample	Low State in Sample
Gene expression	Gene expression probe set	RMA [35] Normalization	Value StepMiner [34] threshold	Value < StepMiner threshold
DNA methylation	CpG site	Scaling by a factor of 10	Value StepMiner threshold	Value < StepMiner threshold
Mutation	Type of mutation per gene	-	Particular type of mutation present in gene	Particular type of mutation absent in gene
Mutation	Mutation per gene	-	Mutation present in gene	Mutation absent in gene
Copy Number	Copy Number amplification per gene	-	Somatic gene amplification present	Somatic gene amplification absent
Copy Number	Copy Number deletion per gene	-	Somatic gene deletion present	Somatic gene deletion absent
Copy Number	Broad Copy Number amplification per segment	-	Broad region amplified	Broad region not amplified
Copy Number	Broad Copy Number deletion per segment	-	Broad Region deleted	Broad Region not deleted

StepMiner [34], which fits patterns of one-step transitions by evaluating every possible placement of the transition (or step) and choosing the one that gives the best fit, was used to derive thresholds that divide the data into low and high states.