Genomic view of samples with 1-5X chromosomes. The X axis shows the markers in a genomic order, with each chromosome uniquely colored. The Y chromosome depicts the measured copy number for each marker in linear scale. The X chromosome is the last chromosome on the right and is shown in orange. (a) A male sample with 1X chromosome. (b) A female sample with 2X chromosomes. (c) A cell line with 3X chromosomes. (d) A cell line with 4X chromosomes. (e) A cell line with 5X chromosomes.

ROC analysis. The x-axis is the rate of false positives (in log₁₀), computed as the proportion of autosomal markers that have copy number below any given threshold (for the 1X calculation). The y-axis depicts sensitivity, defined as the proportion of X chromosome markers that have copy number values below the same threshold (for the 1X calculation). The curve is generated by calculating these values at many different thresholds. The curves from the 3X, 4X, and 5X cell lines were generated in an analogous fashion.

ROC analysis for individual samples. The x-axis is generated in the same fashion as Figure 2, except that the curve for each sample is plotted separately. The average curve is the thick black line.

ROC analysis for allele ratio. The x-axis is the rate of false positives (in log₁₀), computed as the proportion of autosomal markers that have allele ratio above a threshold. The y-axis depicts sensitivity, defined as the proportion of X chromosome markers in the cell line carrying 3X chromosomes that have copy number values below the same threshold. The curve is generated by calculating these values at many different thresholds.

ROC analysis for two-marker smoothing. The same ROC analysis as described in Figure 2 was performed here using the same set of markers (~40 K) as well as using a larger number of markers (~48 K). The ROC analysis was also performed using two-marker smoothing. In this case the smoothing was done for two random markers. If we assume that the performance of individual markers is not correlated with their position (that is, markers close together are likely to have similar performance), then this should be an accurate reflection of the resultant performance with adjacent marker smoothing. We note that at the lower false positive rate for the two-marker smoothed data, the curve is not smooth given low statistics.

Amplification in MCF7. (a) The x-axis shows the markers in a genomic order, with each chromosome uniquely colored. The y-axis depicts the measured copy number for each marker in log₂ (the log scale is used given the high dynamic range). The arrow depicts the position of the locus that was also analyzed by real time PCR. (b) Focused view around the amplification site that was checked with real time PCR. As can be seen, there are several sites of amplifications of different levels. The black bar identifies the region for which average copy number was calculated.

Estimation of copy number of the spikes. The x-axis shows the expected copy number (in log₂) for the individual spiked in PCR fragments, and the y-axis shows the observed copy number for the same spiked in fragments. The linear fit (r² = 0.82) is only for spikes with expected copy number <64 (2⁶) because of the clear saturation above that point.

ACN distributions and reproducibility. (a) Copy number measurements for tumor sample 47 (fairly normal genome content) with genotypes AA colored red, AB colored blue, and BB colored green. (b, c) Allele copy number measurements for tumor 45 (replicate 1). (b) Genotypes derived from replicate 1 are colored AA red, AB blue, and BB green; (c) the genotypes from replicate 2 in the same color scale.

Visualization of individual copy number measurements without matched normal samples. (a-c) Copy number measurements for tumor 48 in genome order from chromosome 1 on the left to chromosome 22 and X on the right. Data are segregated by higher and lower copy number estimates and by homozygosity or heterozygosity. Blue and orange data points are the higher allele copy measurement while green and red data points are the lower copy number measurements. Blue and red data points are homozygous alleles while orange and green are heterozygous alleles. (a) The entire genome. (b) Chromosome 1 through the first 100 Mbp of chromosome 5. (c) Chromosome 1 and the first 50 Mbp of chromosome 2, showing key features of ACN data: an amplification is seen near position 5e7; an extra copy of 1q is seen between ~1.5e8 and 2.5e8; a deletion of 1 copy is seen on the p arm of chromosome 2 between ~2.5e8 and 3e8 (observed in (b) as a complete loss of one copy of chromosome 2). (d) A small section of chromosome 5 from tumor 44. One chromosome is at copy number 0.5 across this region, which indicates a loss of that chromosome. The black arrow shows a region at total copy number 2, which likely includes reduplication of the lost chromosome in the tumor. The red arrow shows a region where both alleles are at copy number 0.5, suggesting a complete deletion. The green arrow shows copy number 1 for the yellow alleles