Genomic instability contributes to tumorigenesis through the amplification and deletion of cancer driver genes. DNA copy number (CN) profiling of ensembles of tumors allows a thermodynamic analysis of the profile for each tumor. The free energy of the distribution of CNs is found to be a monotonically increasing function of the average chromosomal ploidy. The dependence is universal across several cancer types. Surprisal analysis distinguishes two main known subgroups: tumors with cells that have or have not undergone whole-genome duplication (WGD). The analysis uncovers that CN states having a narrower distribution are energetically more favorable toward the WGD transition. Surprisal analysis also determines the deviations from a fully stable-state distribution. These deviations reflect constraints imposed by tumor fitness selection pressures. The results point to CN changes that are more common in high-ploidy tumors and thus support altered selection pressures upon WGD.

基因组不稳定性通过癌症驱动基因的扩增和缺失促进肿瘤发生。肿瘤集合的DNA拷贝数（CN）分析可以对每个肿瘤的概况进行热力学分析。发现CNs分布的自由能是平均染色体倍性的单调递增函数。这种依赖性在几种癌症类型中普遍存在。意外分析区分了两个主要的已知亚组：具有已经或未经历全基因组复制（WGD）的细胞的肿瘤。分析发现，具有较窄分布的CN状态在能量上对WGD过渡更有利。意外分析还确定了与完全稳态分布的偏差。这些偏差反映了肿瘤适应性选择压力所施加的约束。