Meta-analysis of Gene-Level Associations for Rare Variants Based on Single-Variant Statistics

Meta-analysis of genome-wide association studies (GWASs) has led to the discoveries of many common variants associated with complex human diseases. There is a growing recognition that identifying “causal” rare variants also requires large-scale meta-analysis. The fact that association tests with rare variants are performed at the gene level rather than at the variant level poses unprecedented challenges in the meta-analysis. First, different studies may adopt different gene-level tests, so the results are not compatible. Second, gene-level tests require multivariate statistics (i.e., components of the test statistic and their covariance matrix), which are difficult to obtain. To overcome these challenges, we propose to perform gene-level tests for rare variants by combining the results of single-variant analysis (i.e., p values of association tests and effect estimates) from participating studies. This simple strategy is possible because of an insight that multivariate statistics can be recovered from single-variant statistics, together with the correlation matrix of the single-variant test statistics, which can be estimated from one of the participating studies or from a publicly available database. We show both theoretically and numerically that the proposed meta-analysis approach provides accurate control of the type I error and is as powerful as joint analysis of individual participant data. This approach accommodates any disease phenotype and any study design and produces all commonly used gene-level tests. An application to the GWAS summary results of the Genetic Investigation of ANthropometric Traits (GIANT) consortium reveals rare and low-frequency variants associated with human height. The relevant software is freely available.     

Ultraconserved elements in the human genome: association and transmission analyses of highly constrained single-nucleotide polymorphisms

Ultraconserved elements in the human genome likely harbor important biological functions as they are dosage sensitive and are able to direct tissue-specific expression. Because they are under purifying selection, variants in these elements may have a lower frequency in the population but a higher likelihood of association with complex traits. We tested a set of highly constrained SNPs (hcSNPs) distributed genome-wide among ultraconserved and nearly ultraconserved elements for association with seven traits related to reproductive (age at natural menopause, number of children, age at first child, and age at last child) and overall [longevity, body mass index (BMI), and height] fitness. Using up to 24,047 European-American samples from the National Heart, Lung, and Blood Institute Candidate Gene Association Resource (CARe), we observed an excess of associations with BMI and height. In an independent replication panel the most strongly associated SNPs showed an 8.4-fold enrichment of associations at the nominal level, including three variants in previously identified loci and one in a locus (DENND1A) previously shown to be associated with polycystic ovary syndrome. Finally, using 1430 family trios, we showed that the transmissions from heterozygous parents to offspring of the derived alleles of rare (frequency ≤0.5%) hcSNPs are not biased, particularly after adjusting for the rates of genotype missingness and error in the data. The lack of transmission bias ruled out an immediately and strongly deleterious effect due to the rare derived alleles, consistent with the observation that mice homozygous for the deletion of ultraconserved elements showed no overt phenotype. Our study also illustrated the importance of carefully modeling potential technical confounders when analyzing genotype data of rare variants.     

The Severity of Pandemic H1N1 Influenza in the United States,from April to July 2009: A Bayesian Analysis

Background

Accurate measures of the severity of pandemic (H1N1) 2009 influenza (pH1N1) are needed to assess the likely impact of an anticipated resurgence in the autumn in the Northern Hemisphere. Severity has been difficult to measure because jurisdictions with large numbers of deaths and other severe outcomes have had too many cases to assess the total number with confidence. Also, detection of severe cases may be more likely, resulting in overestimation of the severity of an average case. We sought to estimate the probabilities that symptomatic infection would lead to hospitalization, ICU admission, and death by combining data from multiple sources.

Methods and Findings

We used complementary data from two US cities: Milwaukee attempted to identify cases of medically attended infection whether or not they required hospitalization, while New York City focused on the identification of hospitalizations, intensive care admission or mechanical ventilation (hereafter, ICU), and deaths. New York data were used to estimate numerators for ICU and death, and two sources of data—medically attended cases in Milwaukee or self-reported influenza-like illness (ILI) in New York—were used to estimate ratios of symptomatic cases to hospitalizations. Combining these data with estimates of the fraction detected for each level of severity, we estimated the proportion of symptomatic patients who died (symptomatic case-fatality ratio, sCFR), required ICU (sCIR), and required hospitalization (sCHR), overall and by age category. Evidence, prior information, and associated uncertainty were analyzed in a Bayesian evidence synthesis framework. Using medically attended cases and estimates of the proportion of symptomatic cases medically attended, we estimated an sCFR of 0.048% (95% credible interval [CI] 0.026%–0.096%), sCIR of 0.239% (0.134%–0.458%), and sCHR of 1.44% (0.83%–2.64%). Using self-reported ILI, we obtained estimates approximately 7–9× lower. sCFR and sCIR appear to be highest in persons aged 18 y and older, and lowest in children aged 5–17 y. sCHR appears to be lowest in persons aged 5–17; our data were too sparse to allow us to determine the group in which it was the highest.

Conclusions

These estimates suggest that an autumn–winter pandemic wave of pH1N1 with comparable severity per case could lead to a number of deaths in the range from considerably below that associated with seasonal influenza to slightly higher, but with the greatest impact in children aged 0–4 and adults 18–64. These estimates of impact depend on assumptions about total incidence of infection and would be larger if incidence of symptomatic infection were higher or shifted toward adults, if viral virulence increased, or if suboptimal treatment resulted from stress on the health care system; numbers would decrease if the total proportion of the population symptomatically infected were lower than assumed. Please see later in the article for the Editors'' Summary     

Cerebrospinal Fluid PKR Level Predicts Cognitive Decline in Alzheimer’s Disease

The cerebrospinal fluid (CSF) levels of the proapoptotic kinase R (PKR) and its phosphorylated PKR (pPKR) are increased in Alzheimer’s disease (AD), but whether CSF PKR concentrations are associated with cognitive decline in AD patients remain unknown. In this study, 41 consecutive patients with AD and 11 patients with amnestic mild cognitive impairment (aMCI) from our Memory Clinic were included. A lumbar puncture was performed during the following month of the clinical diagnosis and Mini-Mental State Examination (MMSE) evaluations were repeated every 6 months during a mean follow-up of 2 years. In AD patients, linear mixed models adjusted for age and sex were used to assess the cross-sectional and longitudinal associations between MMSE scores and baseline CSF levels of Aβ peptide (Aβ 1-42), Tau, phosphorylated Tau (p-Tau 181), PKR and pPKR. The mean (SD) MMSE at baseline was 20.5 (6.1) and MMSE scores declined over the follow-up (-0.12 point/month, standard error [SE] = 0.03). A lower MMSE at baseline was associated with lower levels of CSF Aβ 1–42 and p-Tau 181/Tau ratio. pPKR level was associated with longitudinal MMSE changes over the follow-up, higher pPKR levels being related with an exacerbated cognitive deterioration. Other CSF biomarkers were not associated with MMSE changes over time. In aMCI patients, mean CSF biomarker levels were not different in patients who converted to AD from those who did not convert.These results suggest that at the time of AD diagnosis, a higher level of CSF pPKR can predict a faster rate of cognitive decline.     

Association between occupational exposure to mineral oil and rheumatoid arthritis: results from the Swedish EIRA case–control study

The aim of the present study was to investigate the association between exposure to mineral oil and the risk of developing rheumatoid arthritis (RA), and in addition to perform a separate analysis on the major subphenotypes for the disease; namely, rheumatoid factor (RF)-positive RA, RF-negative RA, anticitrulline-positive RA and anticitrulline-negative RA, respectively. A population-based case–control study of incident cases of RA was performed among the population aged 18–70 years in a defined area of Sweden during May 1996–December 2003. A case was defined as an individual from the study base who for the first time received a diagnosis of RA according to the American College of Rheumatology criteria of 1987. Controls were randomly selected from the study base with consideration taken for age, gender and residential area. Cases (n = 1,419) and controls (n = 1,674) answered an extensive questionnaire regarding lifestyle factors and occupational exposures, including different types of mineral oils. Sera from cases and controls were investigated for RF and anticitrulline antibodies.     

Risk factors and immunity in a nationally representative population following the 2009 influenza A(H1N1) pandemic

Background

Understanding immunity, incidence and risk factors of the 2009 influenza A(H1N1) pandemic (2009 H1N1) through a national seroprevalence study is necessary for informing public health interventions and disease modelling.

Methods and Findings

We collected 1687 serum samples and individual risk factor data between November-2009 to March-2010, three months after the end of the 2009 H1N1 wave in New Zealand. Participants were randomly sampled from selected general practices countrywide and hospitals in the Auckland region. Baseline immunity was measured from 521 sera collected during 2004 to April-2009. Haemagglutination inhibition (HI) antibody titres of ≥1∶40 against 2009 H1N1 were considered seroprotective as well as seropositive. The overall community seroprevalence was 26.7% (CI:22.6–29.4). The seroprevalence varied across age and ethnicity. Children aged 5–19 years had the highest seroprevalence (46.7%;CI:38.3–55.0), a significant increase from the baseline (14%;CI:7.2–20.8). Older adults aged ≥60 had no significant difference in seroprevalence between the serosurvey (24.8%;CI:18.7–30.9) and baseline (22.6%;CI:15.3–30.0). Pacific peoples had the highest seroprevalence (49.5%;CI:35.1–64.0). There was no significant difference in seroprevalence between both primary (29.6%;CI:22.6–36.5) and secondary healthcare workers (25.3%;CI:20.8–29.8) and community participants. No significant regional variation was observed. Multivariate analysis indicated age as the most important risk factor followed by ethnicity. Previous seasonal influenza vaccination was associated with higher HI titres. Approximately 45.2% of seropositive individuals reported no symptoms.

Conclusions

Based on age and ethnicity standardisation to the New Zealand Population, about 29.5% of New Zealanders had antibody titers at a level consistent with immunity to 2009 H1N1. Around 18.3% of New Zealanders were infected with the virus during the first wave including about one child in every three. Older people were protected due to pre-existing immunity. Age was the most important factor associated with infection followed by ethnicity. Healthcare workers did not appear to have an increased risk of infection compared with the general population.     

Dietary BMAA Exposure in an Amyotrophic Lateral Sclerosis Cluster from Southern France

Background

Dietary exposure to the cyanotoxin BMAA is suspected to be the cause of amyotrophic lateral sclerosis in the Western Pacific Islands. In Europe and North America, this toxin has been identified in the marine environment of amyotrophic lateral sclerosis clusters but, to date, only few dietary exposures have been described.

Objectives

We aimed at identifying cluster(s) of amyotrophic lateral sclerosis in the Hérault district, a coastal district from Southern France, and to search, in the identified area(s), for the existence of a potential dietary source of BMAA.

Methods

A spatio-temporal cluster analysis was performed in the district, considering all incident amyotrophic lateral sclerosis cases identified from 1994 to 2009 by our expert center. We investigated the cluster area with serial collections of oysters and mussels that were subsequently analyzed blind for BMAA concentrations.

Results

We found one significant amyotrophic lateral sclerosis cluster (p = 0.0024), surrounding the Thau lagoon, the most important area of shellfish production and consumption along the French Mediterranean coast. BMAA was identified in mussels (1.8 µg/g to 6.0 µg/g) and oysters (0.6 µg/g to 1.6 µg/g). The highest concentrations of BMAA were measured during summer when the highest picocyanobacteria abundances were recorded.

Conclusions

While it is not possible to ascertain a direct link between shellfish consumption and the existence of this ALS cluster, these results add new data to the potential association of BMAA with sporadic amyotrophic lateral sclerosis, one of the most severe neurodegenerative disorder.     

Antiviral therapy and outcomes of patients with pneumonia caused by influenza A pandemic (H1N1) virus

Background

There is limited data on the clinical outcome of patients with pandemic H1N1 (pH1N1) pneumonia who received oseltamivir treatment, especially when the treatment was administered more than 48 hours after symptom onset.

Methods

During the pandemic in 2009, a cohort of pH1N1 influenza pneumonia was built in China, and their clinical information was collected systematically, and analyzed with Cox models.

Results

920 adults and 541 children with pneumonia who didn''t receive corticosteroids were analyzed. In-hospital mortality was higher in adults who did not receive antiviral therapy (18.2%) than those with who received oseltamivir ≤ 2days (2.9%), between 2–5 days (4.6%) and >5 days after illness onset (4.9%), p<0.01. A similar trend was observed in pediatric patients. Cox regression showed that at 60 days after symptoms onset, 11 patients (10.8%) who did not receive antivirals died versus 4 (1.8%), 18 (3.3%), and 23 (3.7%) patients whose oseltamivir treatment was started ≤ 2days, between 2–5days, and >5 days, respectively. For males patients, aged ≥ 14 years and baseline PaO₂/FiO₂<200, oseltamivir administration reduced the mortality risk by 92.1%, 88% and 83.5%, respectively. Higher doses of oseltamivir (>3.8 mg/kg/d) did not improve clinical outcome (mortality, higher dose 2.5% vs standard dose 2.8%, p>0.05).

Conclusions

Antiviral therapy might reduce mortality of patients with pH1N1 pneumonia, even when initiated more than 48 hours after onset of illness. Greater protective effects might be in males, patients aged 14–60 years, and patients with PaO₂/FiO₂<200.     

A Multi-SNP Locus-Association Method Reveals a Substantial Fraction of the Missing Heritability

There are many known examples of multiple semi-independent associations at individual loci; such associations might arise either because of true allelic heterogeneity or because of imperfect tagging of an unobserved causal variant. This phenomenon is of great importance in monogenic traits but has not yet been systematically investigated and quantified in complex-trait genome-wide association studies (GWASs). Here, we describe a multi-SNP association method that estimates the effect of loci harboring multiple association signals by using GWAS summary statistics. Applying the method to a large anthropometric GWAS meta-analysis (from the Genetic Investigation of Anthropometric Traits consortium study), we show that for height, body mass index (BMI), and waist-to-hip ratio (WHR), 3%, 2%, and 1%, respectively, of additional phenotypic variance can be explained on top of the previously reported 10% (height), 1.5% (BMI), and 1% (WHR). The method also permitted a substantial increase (by up to 50%) in the number of loci that replicate in a discovery-validation design. Specifically, we identified 74 loci at which the multi-SNP, a linear combination of SNPs, explains significantly more variance than does the best individual SNP. A detailed analysis of multi-SNPs shows that most of the additional variability explained is derived from SNPs that are not in linkage disequilibrium with the lead SNP, suggesting a major contribution of allelic heterogeneity to the missing heritability.     

Comparison of 24-h Urinary Aldosterone Level and Random Urinary Aldosterone-to-Creatinine Ratio in the Diagnosis of Primary Aldosteronism

Background

Historically, urinary aldosterone level measurement was a commonly employed confirmatory test to detect primary aldosteronism (PA). However, 24-h urine collection is inconvenient and cumbersome. We hypothesized that random urinary aldosterone measurements with correction for creatinine concentration might be comparable to 24-h urinary aldosterone levels (Uald-24 h) in the diagnosis of PA.

Methods

The non-concurrent prospective study was conducted between June 2006 and March 2008 in patients admitted for confirmation of aldosteronism by salt loading test. A 24-h urine sample, which was collected during hospitalization on the day before saline infusion testing after restoration of serum hypokalemia, was collected from all subjects. Moreover, participants were asked to collect a first bladder voiding random urine sample during clinic visits. Uald-24 h and the random urinary aldosterone-to-creatinine ratio (UACR) were calculated accordingly.

Results

A total of 102 PA patients (71 patients diagnosed of aldosterone-producing adenoma, 31 with idiopathic hyperaldosteronism) and 65 patients with EH were enrolled. The receiver operating characteristic curve showed comparable areas under the curves of UACR and Uald-24 h. The Bland-Altman plot showed mean bias but no obvious heteroscedasticity between the two tests. When using random UACR >3.0 ng/mg creatinine as the cutoff value, we obtained a specificity of 90.6% to confirm PA from essential hypertension.

Conclusions

Our study reinforce that the diagnostic accuracy of random UACR was comparable to that of Uald-24 h in PA patients. With the quickness and simplicity of the UACR method and its equivalence to Uald-24 h, this assay could be a good alternative diagnostic tool for PA confirmation.