Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2011 Nov 20;29(12):1109-13.
doi: 10.1038/nbt.2049.

High order chromatin architecture shapes the landscape of chromosomal alterations in cancer

Geoff Fudenberg  1 , Gad Getz, Matthew Meyerson, Leonid A Mirny
Affiliations

High order chromatin architecture shapes the landscape of chromosomal alterations in cancer

Geoff Fudenberg et al. Nat Biotechnol. .

Abstract

The accumulation of data on structural variation in cancer genomes provides an opportunity to better understand the mechanisms of genomic alterations and the forces of selection that act upon these alterations in cancer. Here we test evidence supporting the influence of two major forces, spatial chromosome structure and purifying (or negative) selection, on the landscape of somatic copy-number alterations (SCNAs) in cancer. Using a maximum likelihood approach, we compare SCNA maps and three-dimensional genome architecture as determined by genome-wide chromosome conformation capture (HiC) and described by the proposed fractal-globule model. This analysis suggests that the distribution of chromosomal alterations in cancer is spatially related to three-dimensional genomic architecture and that purifying selection, as well as positive selection, influences SCNAs during somatic evolution of cancer cells.

PubMed Disclaimer

Figures

Figure 1
Figure 1. 3D proximity as mechanism for SCNA formation
A: Model of how chromosomal architecture and selection can influence observed patterns of somatic copy-number alterations (SCNAs). First spatial proximity of the loop ends makes an SCNA more likely to occur after DNA damage and repair. Next, forces of positive selection and purifying selection act on SCNAs which have arisen, leading to their ultimate fixation or loss. Observed SCNAs in cancer thus reflect both mutational and selective forces. Inset illustrates looping in a simulated fractal globule architecture (coordinates from M. Imakaev). Two contact points are highlighted by spheres and represent potential end-points of SCNAs. B. SCNA length distribution for 60,580 less-recurrent SCNAs (39,071 amplifications, 21,509 deletions) mapped in 3,131 cancer specimens from 26 histological types. Squares show mean number of amplification (red) or deletion (blue) SCNAs after binning at 100 kb resolution (and then averaged over logarithmic intervals). Light magenta lines show ~1/L distributions. Grey line shows the best fit for purifying selection (Eq 4) with a uniform mutation rate. Dark purple line shows best fit for deletions for FG+sel. C: Probability of a contact between two loci distance L apart on a chromosome at 100 kb resolution. The probability is obtained from intra-chromosomal interactions of 22 human chromosomes characterized by the HiC method (human cell line GM06690). Shaded area shows range from 5th and 95th percentiles for number of counts in a 100kb bin at a given distance. The mean contact probability is shown by blue line. Light magenta line shows ~1/L scaling also observed in the fractal globule model of chromatin architecture. Blue dashed line provides a baseline for contact frequency obtained as inter-chromosomal contacts in the same dataset.
Figure 2
Figure 2. Heatmaps for chromosome 17 at 1 Mb resolution
A. SCNA heatmap: the value for site (i,j) is the number of SCNAs starting at genomic location i and ending at location j on the same chromosome. Chromosome band structure from UCSC browser shown on the left side with centromeric bands in red. B. HiC heatmap: site (i,j) has the number of reported interactions between genomic locations i and j at Mb resolution. HiC domain structure is shown on the left side. Domains were determined by thresholding the HiC eigenvector (as in, white represents open domains, dark gray represents closed domains). C. Permuted SCNA heatmap: as in A, but after randomly permuting SCNA locations while keeping SCNA lengths fixed. Visually, the true SCNA heatmap is similar to HiC (Pearson’s r = .55, p <.001, see Supplementary Table S1 for other chromosomes), displaying a “domain” style organization. Cartoons above the heatmaps illustrate how mapped HiC fragments and SCNA end-points can be converted into interactions between genomic locations i and j. Since inter-arm SCNAs, SCNAs with end-points near centromeres or telomeres, and SCNAs < 1Mb were not considered in our statistical analysis, these areas of the heatmaps are grayed out.
Figure 3
Figure 3. Selecting a model of SCNA formation
For each model, the log-likelihood ratio (*BIC-corrected log-likelihood ratio) is shown for the 24,310 observed SCNAs that do not span highly-recurrent SCNA regions listed in. The following six models are considered: Uniform, Uniform+sel, HiC, HiC+sel, FG, FG+sel. HiC model assumes mutation rates proportional to experimentally measured contact probabilities, while FG model assumes mutation rates proportional to mean contact probability in a fractal globule architecture (~1/L). Left y-axis presents BIC-corrected log-likelihood ratio for each model vs. Uniform model. Each model was considered with (+) and without (−) purifying selection. Right y-axis shows the same data as a fold difference in likelihood per cancer specimen (sample) vs. Uniform. Error bars were obtained via bootstrapping: squares represent the median values, bar ends represent the 5th and 95th percentiles. The FG model significantly outperforms other mutational models of SCNA formation, and every model is significantly improved when purifying selection is taken into account.
Figure 4
Figure 4. Permutation analysis of the relationship between SCNAs and megabase-level structure of HiC chromosomal interactions
A. Distribution of log-likelihood ratios for randomly permuted SCNAs given HiC vs. observed SCNAs given HiC over all 22 autosomes. Observed SCNAs (blue arrow) are fit better by HiC contact probability with p<.001. Permutations are performed by shuffling SCNA locations while keeping SCNA lengths fixed. B: Distributions of the same log-likelihood ratios for individual chromosomes (vs their corresponding observed SCNA, blue line). Squares represent median values, error bars respective represent the range from 5th to 25th percentile and 75th to 95th percentile.
See this image and copyright information in PMC

Comment in

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Beroukhim R, et al. The landscape of somatic copy-number alteration across human cancers. Nature. 2010;463:899–905. - PMC - PubMed
    1. Lieberman-Aiden E, et al. Comprehensive mapping of long-range interactions reveals folding principles of the human genome. Science. 2009;326:289–293. - PMC - PubMed
    1. Mirny LA. The fractal globule as a model of chromatin architecture in the cell. Chromosome Research. 2011;19:37–51. - PMC - PubMed
    1. Greenman C, et al. Patterns of somatic mutation in human cancer genomes. Nature. 2007;446:153–158. - PMC - PubMed
    1. Wood LD, et al. The genomic landscapes of human breast and colorectal cancers. Science. 2007;318:1108–1113. - PubMed

Publication types