2. Segmenting Copy Number Data

1. You now need to segment the Copy Number Data to find consensus segments or copy number regions for each sample. To do this, open the newly created Copy Number Data - Sheet 1 (you can close all the other windows to reduce clutter) and select Analysis > CNAM Optimal Segmenting. Make sure both One column per marker and First column of each segment are selected after Segment means output:.

   Due to being an optimal algorithm, CNAM segmenting is compute intensive. To speed up the process, set the number of threads to as many cores as you have on your computer. If hardware acceleration is also available select this option to speed up the process. On the example dataset used for this tutorial (42 samples x 330K markers) it takes about an hour to run on an 8 core machine without hardware acceleration. Using hardware acceleration (NVIDIA Quadro NVS 295) on a 4 core machine it takes about an hour as well. The CNAM Optimal Segmenting window should look like the screenshot in Figure 7. Click Run to begin the segmentation.

   The resulting outputs include Segmentation Run results, a Segment List spreadsheet listing information about all the segments found for every sample, and two segment covariate spreadsheets; Segment Covariates Every Column (better for visualization purposes) and Segmentation Covariates First Column (better for association testing).

   

   Figure 7. CNAM Optimal Segmenting Window