Abstract
Many common diseases have a complex genetic basis in which large numbers of genetic variations combine with environmental and lifestyle factors to determine risk. However, quantifying such polygenic effects and their relationship to disease risk has been challenging. In order to address these difficulties, we developed a global measure of the information content of an individual's genome relative to a reference population, which may be used to assess differences in global genome structure between cases and appropriate controls. Informally this measure, which we call relative genome information (RGI), quantifies the relative "disorder" of an individual's genome. In order to test its ability to predict disease risk, we used RGI to compare single nucleotide polymorphism genotypes from two independent samples of women with early-onset breast cancer with three independent sets of controls. We found that RGI was significantly elevated in both sets of breast cancer cases in comparison with all three sets of controls, with disease risk rising sharply with RGI (odds ratio greater than 12 for the highest percentile RGI). Furthermore, we found that these differences are not due to associations with common variants at a small number of disease-associated loci, but rather are due to the combined associations of thousands of markers distributed throughout the genome. Our results indicate that the information content of an individual's genome may be used to measure the risk of a complex disease, and suggest that early-onset breast cancer has a strongly polygenic basis.
Introduction
The genetic landscape of breast cancer is characterized by a stark dichotomy. On one hand, high-penetrance mutations in genes such as BRCA1 and BRCA2 provide clear risk profiles, yet they account for only a small fraction of total cases. On the other hand, the vast majority of breast cancer cases are thought to be polygenic, arising from the cumulative effect of numerous genetic variations. Traditional Genome-Wide Association Studies (GWAS) have struggled to map this "missing heritability," as individual common variants often carry negligible effect sizes (odds ratios typically less than 1.1).
To bridge this gap, Smyth et al. proposed a shift in perspective: rather than searching for specific "broken" genes, they developed a method to quantify the global "disorder" or "surprisal" of an individual’s genome. This metric, termed Relative Genome Information (RGI), offers a novel lens through which to view disease susceptibility by measuring the aggregate impact of rare variants across the entire genome.
Quantifying Genomic "Surprisal"
The core of this research lies in the application of information theory to genetics. The authors define RGI as the amount of information (measured in bits) required to describe an individual's genotype relative to the allele frequencies of a reference population.
Informally, RGI measures how "unusual" a genome is. An individual carrying mostly common alleles has a highly predictable genome with low RGI. Conversely, an individual carrying an excess of rare alleles—regardless of whether those alleles are known pathogens—possesses a genome with high "surprisal" or disorder. To allow for comparisons across different genotyping platforms and datasets, the researchers utilized Expected Information per Locus (EIL), which normalizes the RGI by the number of Single Nucleotide Polymorphisms (SNPs) analyzed.
Study Design and Methodology
The study evaluated the predictive power of EIL using two independent samples of women with early-onset breast cancer (diagnosed at age 40 or younger):
- The POSH cohort: A large UK-based study of early-onset cases.
- The Australian Breast Cancer Family Study (ABCFS): A population-based case-control-family study.
These cases were compared against three independent control sets, including the WTCCC 1958 British Birth Cohort and the UK National Blood Service (NBS) collection. The analysis focused on approximately 475,000 shared SNPs across the datasets, using logistic generalized additive models to correlate genomic disorder with cancer diagnosis.
Key Findings: The Magnitude of Polygenic Risk
The researchers found that women with early-onset breast cancer exhibited significantly higher EIL compared to all three control groups ($p < 0.01$). This elevation was not localized to a few specific regions but was distributed across 19 of the 22 autosomes, suggesting a systemic, genome-wide phenomenon.
The most striking result was the magnitude of risk associated with high genomic disorder. For individuals in the highest percentile of RGI, the odds ratio for breast cancer exceeded 12. This effect size is comparable to the risk conferred by high-penetrance mutations in BRCA1/2, yet it is derived from the aggregate signal of thousands of variants.
Further analysis confirmed the polygenic nature of this signal. The researchers estimated that the difference in risk at the 99th percentile was driven by the combined associations of approximately 4,954 distinct genetic loci. This confirms that breast cancer susceptibility in these cases is not the result of a few "outlier" SNPs, but rather a global accumulation of rare genetic variations.
Evolutionary and Biological Context
The study also explored how genomic disorder varies across different functional regions of the genome. The lowest EIL values were observed in exonic (protein-coding) regions, while the highest were found in intergenic regions. This pattern aligns with the principles of natural selection: evolutionary pressure is strongest in coding regions to maintain functional stability, whereas non-coding regions are more tolerant of variation.
While the study establishes a strong correlation between RGI and cancer risk, the underlying biological mechanism remains a subject for future research. The authors hypothesize that elevated genomic disorder may serve as a proxy for subtle, germline-wide deficiencies in DNA repair or maintenance, potentially linked to the p53/p63/p73 pathways.
Conclusion and Future Directions
The introduction of Relative Genome Information provides a powerful new tool for assessing complex disease risk. By moving beyond the "one gene, one disease" paradigm, this information-theoretic approach captures the cumulative impact of thousands of variants that traditional GWAS might overlook.
While EIL proved to be a robust predictor of risk, it did not correlate with disease prognosis or age of onset within the case groups, suggesting that the factors driving cancer initiation may differ from those driving its progression. Future work may extend this methodology to other polygenic conditions, such as cardiovascular disease or type 2 diabetes, potentially uncovering the "missing heritability" that has long eluded genomic medicine.
Reference: Smyth, C., et al. "Genome disorder and breast cancer susceptibility." (Page 1-15).
Summary updated