Population structure and eigenanalysis

Current methods for inferring population structure from genetic data do not provide formal significance tests for population differentiation. We discuss an approach to studying population structure (principal components analysis) that was first applied to genetic data by Cavalli-Sforza and colleagues. We place the method on a solid statistical footing, using results from modern statistics to develop formal significance tests. We also uncover a general "phase change" phenomenon about the ability to detect structure in genetic data, which emerges from the statistical theory we use, and has an important implication for the ability to discover structure in genetic data: for a fixed but large dataset size, divergence between two populations (as measured, for example, by a statistic like FST) below a threshold is essentially undetectable, but a little above threshold, detection will be easy. This means that we can predict the dataset size needed to detect structure.

Genetic dissection of complex traits

Medical genetics was revolutionized during the 1980s by the application of genetic mapping to locate the genes responsible for simple Mendelian diseases. Most diseases and traits, however, do not follow simple inheritance patterns. Genetics have thus begun taking up the even greater challenge of the genetic dissection of complex traits. Four major approaches have been developed: linkage analysis, allele-sharing methods, association studies, and polygenic analysis of experimental crosses. This article synthesizes the current state of the genetic dissection of complex traits--describing the methods, limitations, and recent applications to biological problems.

Structural variation of chromosomes in autism spectrum disorder

Structural variation (copy number variation [CNV] including deletion and duplication, translocation, inversion) of chromosomes has been identified in some individuals with autism spectrum disorder (ASD), but the full etiologic role is unknown. We performed genome-wide assessment for structural abnormalities in 427 unrelated ASD cases via single-nucleotide polymorphism microarrays and karyotyping. With microarrays, we discovered 277 unbalanced CNVs in 44% of ASD families not present in 500 controls (and re-examined in another 1152 controls). Karyotyping detected additional balanced changes. Although most variants were inherited, we found a total of 27 cases with de novo alterations, and in three (11%) of these individuals, two or more new variants were observed. De novo CNVs were found in approximately 7% and approximately 2% of idiopathic families having one child, or two or more ASD siblings, respectively. We also detected 13 loci with recurrent/overlapping CNV in unrelated cases, and at these sites, deletions and duplications affecting the same gene(s) in different individuals and sometimes in asymptomatic carriers were also found. Notwithstanding complexities, our results further implicate the SHANK3-NLGN4-NRXN1 postsynaptic density genes and also identify novel loci at DPP6-DPP10-PCDH9 (synapse complex), ANKRD11, DPYD, PTCHD1, 15q24, among others, for a role in ASD susceptibility. Our most compelling result discovered CNV at 16p11.2 (p = 0.002) (with characteristics of a genomic disorder) at approximately 1% frequency. Some of the ASD regions were also common to mental retardation loci. Structural variants were found in sufficiently high frequency influencing ASD to suggest that cytogenetic and microarray analyses be considered in routine clinical workup.

Adjusting multiple testing in multilocus analyses using the eigenvalues of a correlation matrix

Correlated multiple testing is widely performed in genetic research, particularly in multilocus analyses of complex diseases. Failure to control appropriately for the effect of multiple testing will either result in a flood of false-positive claims or in true hits being overlooked. Cheverud proposed the idea of adjusting correlated tests as if they were independent, according to an 'effective number' (M(eff)) of independent tests. However, our experience has indicated that Cheverud's estimate of the Meff is overly large and will lead to excessively conservative results. We propose a more accurate estimate of the M(eff), and design M(eff)-based procedures to control the experiment-wise significant level and the false discovery rate. In an evaluation, based on both real and simulated data, the M(eff)-based procedures were able to control the error rate accurately and consequently resulted in a power increase, especially in multilocus analyses. The results confirm that the M(eff) is a useful concept in the error-rate control of correlated tests. With its efficiency and accuracy, the M(eff) method provides an alternative to computationally intensive methods such as the permutation test.

Joint analysis is more efficient than replication-based analysis for two-stage genome-wide association studies

Genome-wide association is a promising approach to identify common genetic variants that predispose to human disease. Because of the high cost of genotyping hundreds of thousands of markers on thousands of subjects, genome-wide association studies often follow a staged design in which a proportion (pi(samples)) of the available samples are genotyped on a large number of markers in stage 1, and a proportion (pi(samples)) of these markers are later followed up by genotyping them on the remaining samples in stage 2. The standard strategy for analyzing such two-stage data is to view stage 2 as a replication study and focus on findings that reach statistical significance when stage 2 data are considered alone. We demonstrate that the alternative strategy of jointly analyzing the data from both stages almost always results in increased power to detect genetic association, despite the need to use more stringent significance levels, even when effect sizes differ between the two stages. We recommend joint analysis for all two-stage genome-wide association studies, especially when a relatively large proportion of the samples are genotyped in stage 1 (pi(samples) >or= 0.30), and a relatively large proportion of markers are selected for follow-up in stage 2 (pi(markers) >or= 0.01).

A tutorial on statistical methods for population association studies

Although genetic association studies have been with us for many years, even for the simplest analyses there is little consensus on the most appropriate statistical procedures. Here I give an overview of statistical approaches to population association studies, including preliminary analyses (Hardy-Weinberg equilibrium testing, inference of phase and missing data, and SNP tagging), and single-SNP and multipoint tests for association. My goal is to outline the key methods with a brief discussion of problems (population structure and multiple testing), avenues for solutions and some ongoing developments.

Development of human protein reference database as an initial platform for approaching systems biology in humans

Human Protein Reference Database (HPRD) is an object database that integrates a wealth of information relevant to the function of human proteins in health and disease. Data pertaining to thousands of protein-protein interactions, posttranslational modifications, enzyme/substrate relationships, disease associations, tissue expression, and subcellular localization were extracted from the literature for a nonredundant set of 2750 human proteins. Almost all the information was obtained manually by biologists who read and interpreted >300,000 published articles during the annotation process. This database, which has an intuitive query interface allowing easy access to all the features of proteins, was built by using open source technologies and will be freely available at http://www.hprd.org to the academic community. This unified bioinformatics platform will be useful in cataloging and mining the large number of proteomic interactions and alterations that will be discovered in the postgenomic era.

A framework for the interpretation of de novo mutation in human disease

Spontaneously arising (de novo) mutations have an important role in medical genetics. For diseases with extensive locus heterogeneity, such as autism spectrum disorders (ASDs), the signal from de novo mutations is distributed across many genes, making it difficult to distinguish disease-relevant mutations from background variation. Here we provide a statistical framework for the analysis of excesses in de novo mutation per gene and gene set by calibrating a model of de novo mutation. We applied this framework to de novo mutations collected from 1,078 ASD family trios, and, whereas we affirmed a significant role for loss-of-function mutations, we found no excess of de novo loss-of-function mutations in cases with IQ above 100, suggesting that the role of de novo mutations in ASDs might reside in fundamental neurodevelopmental processes. We also used our model to identify ∼1,000 genes that are significantly lacking in functional coding variation in non-ASD samples and are enriched for de novo loss-of-function mutations identified in ASD cases.

Genetic analyses of diverse populations improves discovery for complex traits

Genome-wide association studies (GWAS) have laid the foundation for investigations into the biology of complex traits, drug development and clinical guidelines. However, the majority of discovery efforts are based on data from populations of European ancestry1-3. In light of the differential genetic architecture that is known to exist between populations, bias in representation can exacerbate existing disease and healthcare disparities. Critical variants may be missed if they have a low frequency or are completely absent in European populations, especially as the field shifts its attention towards rare variants, which are more likely to be population-specific4-10. Additionally, effect sizes and their derived risk prediction scores derived in one population may not accurately extrapolate to other populations11,12. Here we demonstrate the value of diverse, multi-ethnic participants in large-scale genomic studies. The Population Architecture using Genomics and Epidemiology (PAGE) study conducted a GWAS of 26 clinical and behavioural phenotypes in 49,839 non-European individuals. Using strategies tailored for analysis of multi-ethnic and admixed populations, we describe a framework for analysing diverse populations, identify 27 novel loci and 38 secondary signals at known loci, as well as replicate 1,444 GWAS catalogue associations across these traits. Our data show evidence of effect-size heterogeneity across ancestries for published GWAS associations, substantial benefits for fine-mapping using diverse cohorts and insights into clinical implications. In the United States-where minority populations have a disproportionately higher burden of chronic conditions13-the lack of representation of diverse populations in genetic research will result in inequitable access to precision medicine for those with the highest burden of disease. We strongly advocate for continued, large genome-wide efforts in diverse populations to maximize genetic discovery and reduce health disparities.

The Evaluation of Genomic Applications in Practice and Prevention (EGAPP) Initiative: methods of the EGAPP Working Group

The Evaluation of Genomic Applications in Practice and Prevention (EGAPP) Initiative, established by the National Office of Public Health Genomics at the Centers for Disease Control and Prevention, supports the development and implementation of a rigorous, evidence-based process for evaluating genetic tests and other genomic applications for clinical and public health practice in the United States. An independent, non-federal EGAPP Working Group (EWG), a multidisciplinary expert panel selects topics, oversees the systematic review of evidence, and makes recommendations based on that evidence. This article describes the EGAPP processes and details the specific methods and approaches used by the EWG.

The use of a genetic map of biallelic markers in linkage studies

Improvements in genetic mapping techniques have driven recent progress in human genetics. The use of single nucleotide polymorphisms (SNPs) as biallelic genetic markers offers the promise of rapid, highly automated genotyping. As maps of SNPs and the techniques for genotyping them are being developed, it is important to consider what properties such maps must have in order for them to be useful for linkage studies. I examine how polymorphic and densely spaced biallelic markers need to be for extraction of most of the inheritance information from human pedigrees, and compare maps of biallelics with today's genome-scanning sets of microsatellite markers. I conclude that a map of 700-900 moderately polymorphic biallelic markers is equivalent--and a map of 1,500-3,000 superior--to the current 300-400 microsatellite marker sets.

Managing incidental findings and research results in genomic research involving biobanks and archived data sets

Biobanks and archived data sets collecting samples and data have become crucial engines of genetic and genomic research. Unresolved, however, is what responsibilities biobanks should shoulder to manage incidental findings and individual research results of potential health, reproductive, or personal importance to individual contributors (using "biobank" here to refer both to collections of samples and collections of data). This article reports recommendations from a 2-year project funded by the National Institutes of Health. We analyze the responsibilities involved in managing the return of incidental findings and individual research results in a biobank research system (primary research or collection sites, the biobank itself, and secondary research sites). We suggest that biobanks shoulder significant responsibility for seeing that the biobank research system addresses the return question explicitly. When reidentification of individual contributors is possible, the biobank should work to enable the biobank research system to discharge four core responsibilities to (1) clarify the criteria for evaluating findings and the roster of returnable findings, (2) analyze a particular finding in relation to this, (3) reidentify the individual contributor, and (4) recontact the contributor to offer the finding. We suggest that findings that are analytically valid, reveal an established and substantial risk of a serious health condition, and are clinically actionable should generally be offered to consenting contributors. This article specifies 10 concrete recommendations, addressing new biobanks as well as those already in existence.

Genetic Evaluation of Cardiomyopathy-A Heart Failure Society of America Practice Guideline

This guideline describes the approach and expertise needed for the genetic evaluation of cardiomyopathy. First published in 2009 by the Heart Failure Society of America (HFSA), the guideline has now been updated in collaboration with the American College of Medical Genetics and Genomics (ACMG). The writing group, composed of cardiologists and genetics professionals with expertise in adult and pediatric cardiomyopathy, reflects the emergence and increased clinical activity devoted to cardiovascular genetic medicine. The genetic evaluation of cardiomyopathy is a rapidly emerging key clinical priority, because high-throughput sequencing is now feasible for clinical testing and conventional interventions can improve survival, reduce morbidity, and enhance quality of life. Moreover, specific interventions may be guided by genetic analysis. A systematic approach is recommended: always a comprehensive family history; an expert phenotypic evaluation of the proband and at-risk family members to confirm a diagnosis and guide genetic test selection and interpretation; referral to expert centers as needed; genetic testing, with pre- and post-test genetic counseling; and specific guidance as indicated for drug and device therapies. The evaluation of infants and children demands special expertise. The approach to managing secondary and incidental sequence findings as recommended by the ACMG is provided.

Integrating Genomics into Healthcare: A Global Responsibility

Genomic sequencing is rapidly transitioning into clinical practice, and implementation into healthcare systems has been supported by substantial government investment, totaling over US$4 billion, in at least 14 countries. These national genomic-medicine initiatives are driving transformative change under real-life conditions while simultaneously addressing barriers to implementation and gathering evidence for wider adoption. We review the diversity of approaches and current progress made by national genomic-medicine initiatives in the UK, France, Australia, and US and provide a roadmap for sharing strategies, standards, and data internationally to accelerate implementation.

Genetic relatedness analysis: modern data and new challenges

Individuals who belong to the same family or the same population are related because of their shared ancestry. Population and quantitative genetics theory is built with parameters that describe relatedness, and the estimation of these parameters from genetic markers enables progress in fields as disparate as plant breeding, human disease gene mapping and forensic science. The large number of multiallelic microsatellite loci and biallelic SNPs that are now available have markedly increased the precision with which relationships can be estimated, although they have also revealed unexpected levels of genomic heterogeneity of relationship measures.

Gene-environment interaction: definitions and study designs

Study of gene-environment interaction is important for improving accuracy and precision in the assessment of both genetic and environmental influences. This overview presents a simple definition of gene-environment interaction and suggests study designs for detecting it. Gene-environment interaction is defined as "a different effect of an environmental exposure on disease risk in persons with different genotypes," or, alternatively, "a different effect of a genotype on disease risk in persons with different environmental exposures." Under this strictly statistical definition, the presence or absence of interaction depends upon the scale of measurement (additive or multiplicative). The decision of which scale is appropriate will be governed by many factors, including the main objective of an investigation (discovery of etiology, public health prediction, etc.) and the hypothesized pathophysiologic model. Five biologically plausible models are described for the relations between genotypes and environmental exposures, in terms of their effects on disease risk. Each of these models leads to a different set of predictions about disease risk in individuals classified by presence or absence of a high-risk genotype and environmental exposure. Classification according to the exposure is relatively easy, using conventional epidemiologic methods. Classification according to the high-risk genotype is more difficult, but several alternative strategies are suggested.

Emerging clinical applications of RNA

RNA is a versatile biological macromolecule that is crucial in mobilizing and interpreting our genetic information. It is not surprising then that researchers have sought to exploit the inherent properties of RNAs so as to interfere with or repair dysfunctional nucleic acids or proteins and to stimulate the production of therapeutic gene products in a variety of pathological situations. The first generation of the resulting RNA therapeutics are now being evaluated in clinical trials, raising significant interest in this emerging area of medical research.

Artificial intelligence in clinical and genomic diagnostics

Artificial intelligence (AI) is the development of computer systems that are able to perform tasks that normally require human intelligence. Advances in AI software and hardware, especially deep learning algorithms and the graphics processing units (GPUs) that power their training, have led to a recent and rapidly increasing interest in medical AI applications. In clinical diagnostics, AI-based computer vision approaches are poised to revolutionize image-based diagnostics, while other AI subtypes have begun to show similar promise in various diagnostic modalities. In some areas, such as clinical genomics, a specific type of AI algorithm known as deep learning is used to process large and complex genomic datasets. In this review, we first summarize the main classes of problems that AI systems are well suited to solve and describe the clinical diagnostic tasks that benefit from these solutions. Next, we focus on emerging methods for specific tasks in clinical genomics, including variant calling, genome annotation and variant classification, and phenotype-to-genotype correspondence. Finally, we end with a discussion on the future potential of AI in individualized medicine applications, especially for risk prediction in common complex diseases, and the challenges, limitations, and biases that must be carefully addressed for the successful deployment of AI in medical applications, particularly those utilizing human genetics and genomics data.

Joint analysis of psychiatric disorders increases accuracy of risk prediction for schizophrenia, bipolar disorder, and major depressive disorder

Genetic risk prediction has several potential applications in medical research and clinical practice and could be used, for example, to stratify a heterogeneous population of patients by their predicted genetic risk. However, for polygenic traits, such as psychiatric disorders, the accuracy of risk prediction is low. Here we use a multivariate linear mixed model and apply multi-trait genomic best linear unbiased prediction for genetic risk prediction. This method exploits correlations between disorders and simultaneously evaluates individual risk for each disorder. We show that the multivariate approach significantly increases the prediction accuracy for schizophrenia, bipolar disorder, and major depressive disorder in the discovery as well as in independent validation datasets. By grouping SNPs based on genome annotation and fitting multiple random effects, we show that the prediction accuracy could be further improved. The gain in prediction accuracy of the multivariate approach is equivalent to an increase in sample size of 34% for schizophrenia, 68% for bipolar disorder, and 76% for major depressive disorders using single trait models. Because our approach can be readily applied to any number of GWAS datasets of correlated traits, it is a flexible and powerful tool to maximize prediction accuracy. With current sample size, risk predictors are not useful in a clinical setting but already are a valuable research tool, for example in experimental designs comparing cases with high and low polygenic risk.

Genetics, statistics and human disease: analytical retooling for complexity

Molecular biologists and geneticists alike now acknowledge that most common human diseases with a genetic component are likely to have complex etiologies. Yet despite this belief, many statistical geneticists continue applying, in slightly new and different ways, methodologies that were developed to dissect much simpler etiologies. In this article, we characterize, with examples, the various factors that can complicate genetic analysis and demonstrate their shared features and how they affect genetic analysis. We describe a variety of approaches that are currently available, revealing methodological gaps and suggesting new directions for method development. Finally, we propose a comprehensive two-step approach to analysis that systemically addresses the different genetic factors that are likely to underlie complex diseases.

The case for a US prospective cohort study of genes and environment

Information from the Human Genome Project will be vital for defining the genetic and environmental factors that contribute to health and disease. Well-designed case-control studies of people with and without a particular disease are essential for this, but rigorous and unbiased conclusions about the causes of diseases and their population-wide impact will require a representative population to be monitored over time (a prospective cohort study). The time is right for the United States to consider such a project.

Society for Endocrinology UK guidance on the initial evaluation of an infant or an adolescent with a suspected disorder of sex development (Revised 2015)

It is paramount that any child or adolescent with a suspected disorder of sex development (DSD) is assessed by an experienced clinician with adequate knowledge about the range of conditions associated with DSD. If there is any doubt, the case should be discussed with the regional DSD team. In most cases, particularly in the case of the newborn, the paediatric endocrinologist within the regional team acts commonly as the first point of contact. This clinician should be part of a multidisciplinary team experienced in management of DSD and should ensure that the affected person and parents have access to specialist psychological support and that their information needs are comprehensively addressed. The underlying pathophysiology of DSD and the strengths and weaknesses of the tests that can be performed should be discussed with the parents and affected young person and tests undertaken in a timely fashion. Finally, in the field of rare conditions, it is imperative that the clinician shares the experience with others through national and international clinical and research collaboration.

Opportunities and challenges for the integration of massively parallel genomic sequencing into clinical practice: lessons from the ClinSeq project

PURPOSE

The debate surrounding the return of results from high-throughput genomic interrogation encompasses many important issues including ethics, law, economics, and social policy. As well, the debate is also informed by the molecular, genetic, and clinical foundations of the emerging field of clinical genomics, which is based on this new technology. This article outlines the main biomedical considerations of sequencing technologies and demonstrates some of the early clinical experiences with the technology to enable the debate to stay focused on real-world practicalities.

METHODS

These experiences are based on early data from the ClinSeq project, which is a project to pilot the use of massively parallel sequencing in a clinical research context with a major aim to develop modes of returning results to individual subjects.

RESULTS

The study has enrolled >900 subjects and generated exome sequence data on 572 subjects. These data are beginning to be interpreted and returned to the subjects, which provides examples of the potential usefulness and pitfalls of clinical genomics.

CONCLUSION

There are numerous genetic results that can be readily derived from a genome including rare, high-penetrance traits, and carrier states. However, much work needs to be done to develop the tools and resources for genomic interpretation. The main lesson learned is that a genome sequence may be better considered as a health-care resource, rather than a test, one that can be interpreted and used over the lifetime of the patient.

Strategies for studying heterogeneous genetic traits in humans by using a linkage map of restriction fragment length polymorphisms

Simple single-gene disorders in humans can be genetically mapped by using traditional methods of linkage analysis and increasingly abundant restriction fragment length polymorphisms (RFLPs). Many human diseases and traits, however, can be expected to be genetically heterogeneous (i.e., caused by any one of several genes), and traditional linkage analysis is much less effective in such circumstances. We present two methods, interval mapping and simultaneous search, designed to exploit the full power of a linkage map of the DNA markers. For the simplest situations, only 1/3 as many affected families are needed to map a heterogeneous trait by using these methods. Only 1/5-1/50 as many are needed to detect that genetic heterogeneity is present.

Guidelines for genotyping in genomewide linkage studies: single-nucleotide-polymorphism maps versus microsatellite maps

Genomewide linkage scans have traditionally employed panels of microsatellite markers spaced at intervals of approximately 10 cM across the genome. However, there is a growing realization that a map of closely spaced single-nucleotide polymorphisms (SNPs) may offer equal or superior power to detect linkage, compared with low-density microsatellite maps. We performed a series of simulations to calculate the information content associated with microsatellite and SNP maps across a range of different marker densities and heterozygosities for sib pairs (with and without parental genotypes), sib trios, and sib quads. In the case of microsatellite markers, we varied density across 11 levels (1 marker every 0.5, 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 cM) and marker heterozygosity across 6 levels (2, 3, 4, 5, 10, or 20 equally frequent alleles), whereas, in the case of SNPs, we varied marker density across 4 levels (1 marker every 0.1, 0.2, 0.5, or 1 cM) and minor-allele frequency across 7 levels (0.5, 0.4, 0.3, 0.2, 0.1, 0.05, and 0.01). When parental genotypes were available, a map consisting of microsatellites spaced every 2 cM or a relatively sparse map of SNPs (i.e., at least 1 SNP/cM) was sufficient to extract most of the inheritance information from the map (>95% in most cases). However, when parental genotypes were unavailable, it was important to use as dense a map of markers as possible to extract the greatest amount of inheritance information. It is important to note that the information content associated with a traditional map of microsatellite markers (i.e., 1 marker every ~10 cM) was significantly lower than the information content associated with a dense map of SNPs or microsatellites. These results strongly suggest that previous linkage studies that employed sparse microsatellite maps could benefit substantially from reanalysis by use of a denser map of markers.