RNASeq Gene Expression Calculation Methodology

Data retrieved from The Cancer Genome Atlas

Normalization
TCGA's RNASeq V2, level 3 data is used exclusively. Data normalized by TCGA is used directly. The values are upper quartile normalized RSEM count estimates for sequence reads mapping to specific genes (RSEM are detailed in this article [PMID: 20802226]

Fold Change
For each gene and each patient, we divide the RSEM value for the tumor sample by the RSEM value for the control sample: Fold Change = tumor/control Where tumor is the RSEM count for gene X in the patient's tumor sample and control is the same for their control sample. This 'fold change' is then log2 transformed: Log2(Fold Change) Under-expressed genes will have negative log2 values; over-expressed gene will have positive log 2 values.

Fold Change

For each gene and each patient, we divide the RSEM value for the tumor sample by the RSEM value for the control sample:

Fold Change = tumor/control

Where tumor is the RSEM count for gene X in the patient's tumor sample and control is the same for their control sample.
This 'fold change' is then log2 transformed: Log2(Fold Change)
Under-expressed genes will have negative log2 values; over-expressed gene will have positive log 2 values.

RNASeq Gene Expression Calculation Methodology Data retrieved from The Cancer Genome Atlas

RNASeq Gene Expression Calculation Methodology

Data retrieved from The Cancer Genome Atlas