Hi everyone,
It is very nice to use the chip intensity data to estimate the FDR of the discovered CNVs. But the algorithm implemented in the GenomeStrip is difficult for me to understand when a CNVR contains many probes.
You say that using the ranks at each probe, the samples are re-ranked across all probes. It is difficult for me to understand the re-ranked procedures. Is this according to the sum of the ranks for these probes for each sample?
If I have a set of CNV and have also been genotyped, more, I also have the normalized intensity for each probe for each sample, how should I use this methods to estimate the accuracy of this data.
Many thanks. Any suggestion will be appreciated!
Zheng Zhuqing