CNV data for Urothelial Bladder carcinoma

Question

[arpa530]

Dear all,

Presently I am working on the CNV data analysis of urothelial bladder carcinoma. I have basically downloaded the data from UCSC cancer genome browser. There are five data sets available for CNV study.The genomic matrix is a tab separated file containing the genomic values for each sample for each probe. The matrix are arranged so that the columns are samples and rows are probes. In addition to this there is one such value after the genomic coordinate. I am wondering what is that value? Are they P-value, Z-value or Q-value. I really didn't get that. Please help me to understand this. I have pasted few examples here.

Example:

TCGA-HT-A74H-01  chr1  3208470   3746015   -0.0133
TCGA-HT-A74H-01  chr1  3752621   6231496   -0.358
TCGA-HT-A74H-01  chr1  6237409   7138986   0.007
TCGA-HT-A74H-01  chr1  7139130   7563206   -0.3678
TCGA-HT-A74H-01  chr1  7563975   46755042  0.0016
TCGA-HT-A74H-01  chr1  46757779  48117532  -0.3615

Answer 1

[Sean Davis]

They represent log2 ratios of tumor copy number vs. reference. Two copies in the tumor would be approximately 0, while a one-copy loss would be approximately -1 (if there is no admixture, etc.).

[arpa530]

Dear Davis,

Thanks for your response. I have gone though your answer and now I am quite relaxed to get this answer. As it is a log2 ratio that means should I perform normalization the data? Please let me know. I have basically taken the four CNV data set from UCSC cancer genome browser to perform metanalysis.

[Sean Davis]

These are not the "raw data". Instead, these data have already undergone normalization and "segmentation", I suspect. You'll need to refer to the source of the data for more details.

[arpa530]

Dear Davis,

Thanks for your kind response. I will surely do that.