Deepti Gurdasani

A General Approach for Haplotype Phasing across the Full Spectrum of Relatedness

Many existing cohorts contain a range of relatedness between genotyped individuals, either by design or by chance. Haplotype estimation in such cohorts is a central step in many downstream analyses. Using genotypes from six cohorts from isolated populations and two cohorts from non-isolated populations, we have investigated the performance of different phasing methods designed for nominally ‘unrelated’ individuals. We find that SHAPEIT2 produces much lower switch error rates in all cohorts compared to other methods, including those designed specifically for isolated populations. In particular, when large amounts of IBD sharing is present, SHAPEIT2 infers close to perfect haplotypes. Based on these results we have developed a general strategy for phasing cohorts with any level of implicit or explicit relatedness between individuals. First SHAPEIT2 is run ignoring all explicit family information. We then apply a novel HMM method (duoHMM) to combine the SHAPEIT2 haplotypes with any family information to infer the inheritance pattern of each meiosis at all sites across each chromosome. This allows the correction of switch errors, detection of recombination events and genotyping errors. We show that the method detects numbers of recombination events that align very well with expectations based on genetic maps, and that it infers far fewer spurious recombination events than Merlin. The method can also detect genotyping errors and infer recombination events in otherwise uninformative families, such as trios and duos. The detected recombination events can be used in association scans for recombination phenotypes. The method provides a simple and unified approach to haplotype estimation, that will be of interest to researchers in the fields of human, animal and plant genetics.

The African Genome Variation Project shapes medical genetics in Africa

Given the importance of Africa to studies of human origins and disease susceptibility, detailed characterization of African genetic diversity is needed. The African Genome Variation Project provides a resource with which to design, implement and interpret genomic studies in sub-Saharan Africa and worldwide. The African Genome Variation Project represents dense genotypes from 1,481 individuals and whole-genome sequences from 320 individuals across sub-Saharan Africa. Using this resource, we find novel evidence of complex, regionally distinct hunter-gatherer and Eurasian admixture across sub-Saharan Africa. We identify new loci under selection, including loci related to malaria susceptibility and hypertension. We show that modern imputation panels (sets of reference genotypes from which unobserved or missing genotypes in study sets can be inferred) can identify association signals at highly differentiated loci across populations in sub-Saharan Africa. Using whole-genome sequencing, we demonstrate further improvements in imputation accuracy, strengthening the case for large-scale sequencing efforts of diverse African haplotypes. Finally, we present an efficient genotype array design capturing common genetic variation in Africa.

Association of HIV and ART with cardiometabolic traits in sub-Saharan Africa: a systematic review and meta-analysis

Background Sub-Saharan Africa (SSA) has the highest burden of HIV in the world and a rising prevalence of cardiometabolic disease; however, the interrelationship between HIV, antiretroviral therapy (ART) and cardiometabolic traits is not well described in SSA populations. Methods We conducted a systematic review and meta-analysis through MEDLINE and EMBASE (up to January 2012), as well as direct author contact. Eligible studies provided summary or individual-level data on one or more of the following traits in HIV+ and HIV-, or ART+ and ART- subgroups in SSA: body mass index (BMI), systolic blood pressure (SBP), diastolic blood pressure (DBP), high-density lipoprotein (HDL), low-density lipoprotein (LDL), triglycerides (TGs) and fasting blood glucose (FBG) or glycated hemoglobin (HbA1c). Information was synthesized under a random-effects model and the primary outcomes were the standardized mean differences (SMD) of the specified traits between subgroups of participants. Results Data were obtained from 49 published and 3 unpublished studies which reported on 29 755 individuals. HIV infection was associated with higher TGs [SMD, 0.26; 95% confidence interval (CI), 0.08 to 0.44] and lower HDL (SMD, −0.59; 95% CI, −0.86 to −0.31), BMI (SMD, −0.32; 95% CI, −0.45 to −0.18), SBP (SMD, −0.40; 95% CI, −0.55 to −0.25) and DBP (SMD, −0.34; 95% CI, −0.51 to −0.17). Among HIV+ individuals, ART use was associated with higher LDL (SMD, 0.43; 95% CI, 0.14 to 0.72) and HDL (SMD, 0.39; 95% CI, 0.11 to 0.66), and lower HbA1c (SMD, −0.34; 95% CI, −0.62 to −0.06). Fully adjusted estimates from analyses of individual participant data were consistent with meta-analysis of summary estimates for most traits. Conclusions Broadly consistent with results from populations of European descent, these results suggest differences in cardiometabolic traits between HIV-infected and uninfected individuals in SSA, which might be modified by ART use. In a region with the highest burden of HIV, it will be important to clarify these findings to reliably assess the need for monitoring and managing cardiometabolic risk in HIV-infected populations in SSA.

Tracing the Route of Modern Humans out of Africa by Using 225 Human Genome Sequences from Ethiopians and Egyptians

The predominantly African origin of all modern human populations is well established, but the route taken out of Africa is still unclear. Two alternative routes, via Egypt and Sinai or across the Bab el Mandeb strait into Arabia, have traditionally been proposed as feasible gateways in light of geographic, paleoclimatic, archaeological, and genetic evidence. Distinguishing among these alternatives has been difficult. We generated 225 whole-genome sequences (225 at 8× depth, of which 8 were increased to 30×; Illumina HiSeq 2000) from six modern Northeast African populations (100 Egyptians and five Ethiopian populations each represented by 25 individuals). West Eurasian components were masked out, and the remaining African haplotypes were compared with a panel of sub-Saharan African and non-African genomes. We showed that masked Northeast African haplotypes overall were more similar to non-African haplotypes and more frequently present outside Africa than were any sets of haplotypes derived from a West African population. Furthermore, the masked Egyptian haplotypes showed these properties more markedly than the masked Ethiopian haplotypes, pointing to Egypt as the more likely gateway in the exodus to the rest of the world. Using five Ethiopian and three Egyptian high-coverage masked genomes and the multiple sequentially Markovian coalescent (MSMC) approach, we estimated the genetic split times of Egyptians and Ethiopians from non-African populations at 55,000 and 65,000 years ago, respectively, whereas that of West Africans was estimated to be 75,000 years ago. Both the haplotype and MSMC analyses thus suggest a predominant northern route out of Africa via Egypt.

Lipoprotein(a) and Risk of Coronary, Cerebrovascular, and Peripheral Artery Disease The EPIC-Norfolk Prospective Population Study

Objective—Although the association between circulating levels of lipoprotein(a) [Lp(a)] and risk of coronary artery disease (CAD) and stroke is well established, its role in risk of peripheral arterial disease (PAD) remains unclear. Here, we examine the association between Lp(a) levels and PAD in a large prospective cohort. To contextualize these findings, we also examined the association between Lp(a) levels and risk of stroke and CAD and studied the role of low-density lipoprotein as an effect modifier of Lp(a)-associated cardiovascular risk. Methods and Results—Lp(a) levels were measured in apparently healthy participants in the European Prospective Investigation of Cancer (EPIC)-Norfolk cohort. Cox regression was used to quantify the association between Lp(a) levels and risk of PAD, stroke, and CAD outcomes. During 212 981 person-years at risk, a total of 2365 CAD, 284 ischemic stroke, and 596 PAD events occurred in 18 720 participants. Lp(a) was associated with PAD and CAD outcomes but not with ischemic stroke (hazard ratio per 2.7-fold increase in Lp(a) of 1.37, 95% CI 1.25–1.50, 1.13, 95% CI 1.04–1.22 and 0.91, 95% CI 0.79–1.03, respectively). Low-density lipoprotein cholesterol levels did not modify these associations. Conclusion—Lp(a) levels were associated with future PAD and CAD events. The association between Lp(a) and cardiovascular disease was not modified by low-density lipoprotein cholesterol levels.

Polymorphisms of large effect explain the majority of the host genetic contribution to variation of HIV-1 virus load

Significance A proportion of the variation in HIV-1 viral load in the infected population is influenced by host genetics. Using a large sample of infected individuals (n = 6,315) with genome-wide genotype data, we sought to map genomic regions that influence HIV viral load and quantify their impact. We identified amino acid positions located in the binding groove of class I HLA proteins (HLA-A and -B) and SNPs in the chemokine (C-C motif) receptor 5 gene region that together explain 14.5% of the observed variation in HIV viral load. Controlling for these signals, we estimated that an additional 5.5% can be explained by common, additive genetic variation. Thus, we demonstrate that common variants of large effect explain the majority of the host genetic component of HIV viral load. Previous genome-wide association studies (GWAS) of HIV-1–infected populations have been underpowered to detect common variants with moderate impact on disease outcome and have not assessed the phenotypic variance explained by genome-wide additive effects. By combining the majority of available genome-wide genotyping data in HIV-infected populations, we tested for association between ∼8 million variants and viral load (HIV RNA copies per milliliter of plasma) in 6,315 individuals of European ancestry. The strongest signal of association was observed in the HLA class I region that was fully explained by independent effects mapping to five variable amino acid positions in the peptide binding grooves of the HLA-B and HLA-A proteins. We observed a second genome-wide significant association signal in the chemokine (C-C motif) receptor (CCR) gene cluster on chromosome 3. Conditional analysis showed that this signal could not be fully attributed to the known protective CCR5Δ32 allele and the risk P1 haplotype, suggesting further causal variants in this region. Heritability analysis demonstrated that common human genetic variation—mostly in the HLA and CCR5 regions—explains 25% of the variability in viral load. This study suggests that analyses in non-European populations and of variant classes not assessed by GWAS should be priorities for the field going forward.

The Association Between Circulating Lipoprotein(a) and Type 2 Diabetes: Is It Causal?

Epidemiological evidence supports a direct and causal association between lipoprotein(a) [Lp(a)] levels and coronary risk, but the nature of the association between Lp(a) levels and risk of type 2 diabetes (T2D) is unclear. In this study, we assessed the association of Lp(a) levels with risk of incident T2D and tested whether Lp(a) levels are causally linked to T2D. We analyzed data on 18,490 participants from the European Prospective Investigation of Cancer (EPIC)-Norfolk cohort that included adults aged 40–79 years at baseline 1993–1997. During an average 10 years of follow-up, 593 participants developed incident T2D. Cox regression models were used to estimate the association between Lp(a) levels and T2D. In Mendelian randomization analyses, based on EPIC-Norfolk combined with DIAbetes Genetics Replication And Meta-analysis data involving a total of 10,088 diabetes case participants and 68,346 control participants, we used a genetic variant (rs10455872) as an instrument to test whether the association between Lp(a) levels and T2D is causal. In adjusted analyses, there was an inverse association between Lp(a) levels and T2D: hazard ratio was 0.63 (95% CI 0.49–0.81; P trend = 0.003) comparing the top versus bottom quintile of Lp(a). In EPIC-Norfolk, a 1-SD increase in logLp(a) was associated with a lower risk of T2D (odds ratio [OR] 0.88 [95% CI: 0.80–0.95]). However, in Mendelian randomization analyses, a 1-SD increase in logLp(a) due to rs10455872, which explained 26.8% of the variability in Lp(a) levels, was not associated with risk of T2D (OR 1.03 [0.96–1.10]; P = 0.41). These prospective findings demonstrate a strong inverse association of Lp(a) levels with risk of T2D. However, a genetic variant that elevated Lp(a) levels was not associated with risk of T2D, suggesting that elevated Lp(a) levels are not causally associated with a lower risk of T2D.

An evaluation of HIV elite controller definitions within a large seroconverter cohort collaboration.

Background Understanding the mechanisms underlying viral control is highly relevant to vaccine studies and elite control (EC) of HIV infection. Although numerous definitions of EC exist, it is not clear which, if any, best identify this rare phenotype. Methods We assessed a number of EC definitions used in the literature using CASCADE data of 25,692 HIV seroconverters. We estimated proportions maintaining EC of total ART-naïve follow-up time, and disease progression, comparing to non-EC. We also examined HIV-RNA and CD4 values and CD4 slope during EC and beyond (while ART naïve). Results Most definitions classify ∼1% as ECs with median HIV-RNA 43–903 copies/ml and median CD4>500 cells/mm3. Beyond EC status, median HIV-RNA levels remained low, although often detectable, and CD4 values high but with strong evidence of decline for all definitions. Median % ART-naïve time as EC was ≥92% although overlap between definitions was low. EC definitions with consecutive HIV-RNA measurements <75 copies/ml with follow-up≥ six months, or with 90% of measurements <400 copies/ml over ≥10 year follow-up preformed best overall. Individuals thus defined were less likely to progress to endpoint (hazard ratios ranged from 12.5–19.0 for non-ECs compared to ECs). Conclusions ECs are rare, less likely to progress to clinical disease, but may eventually lose control. We suggest definitions requiring individuals to have consecutive undetectable HIV-RNA measurements for ≥ six months or otherwise with >90% of measurements <400 copies/ml over ≥10 years be used to define this phenotype.

A systematic review of definitions of extreme phenotypes of HIV control and progression

The study of individuals at opposite ends of the HIV clinical spectrum can provide invaluable insights into HIV biology. Heterogeneity in criteria used to define these individuals can introduce inconsistencies in results from research and make it difficult to identify biological mechanisms underlying these phenotypes. In this systematic review, we formally quantified the heterogeneity in definitions used for terms referring to extreme phenotypes in the literature, and identified common definitions and components used to describe these phenotypes. We assessed 714 definitions of HIV extreme phenotypes in 501 eligible studies published between 1 January 2000 and 15 March 2012, and identified substantial variation among these. This heterogeneity in definitions may represent important differences in biological endophenotypes and clinical progression profiles of individuals selected by these, suggesting the need for harmonized definitions. In this context, we were able to identify common components in existing definitions that may provide a framework for developing consensus definitions for these phenotypes in HIV infection.

Linear mixed model for heritability estimation that explicitly addresses environmental variation

The linear mixed model (LMM) is now routinely used to estimate heritability. Unfortunately, as we demonstrate, LMM estimates of heritability can be inflated when using a standard model. To help reduce this inflation, we used a more general LMM with two random effects—one based on genomic variants and one based on easily measured spatial location as a proxy for environmental effects. We investigated this approach with simulated data and with data from a Uganda cohort of 4,778 individuals for 34 phenotypes including anthropometric indices, blood factors, glycemic control, blood pressure, lipid tests, and liver function tests. For the genomic random effect, we used identity-by-descent estimates from accurately phased genome-wide data. For the environmental random effect, we constructed a covariance matrix based on a Gaussian radial basis function. Across the simulated and Ugandan data, narrow-sense heritability estimates were lower using the more general model. Thus, our approach addresses, in part, the issue of “missing heritability” in the sense that much of the heritability previously thought to be missing was fictional. Software is available at https://github.com/MicrosoftGenomics/FaST-LMM.