Testing for association with rare variants in the coding and non-coding genome: RAVA-FIRST, a new approach based on CADD deleteriousness score.

Rare variant association tests (RVAT) have been developed to study the contribution of rare variants widely accessible through high-throughput sequencing technologies. RVAT require to aggregate rare variants in testing units and to filter variants to retain only the most likely causal ones. In the exome, genes are natural testing units and variants are usually filtered based on their functional consequences. However, when dealing with whole-genome sequence (WGS) data, both steps are challenging. No natural biological unit is available for aggregating rare variants. Sliding windows procedures have been proposed to circumvent this difficulty, however they are blind to biological information and result in a large number of tests. We propose a new strategy to perform RVAT on WGS data: "RAVA-FIRST" (RAre Variant Association using Functionally-InfoRmed STeps) comprising three steps. (1) New testing units are defined genome-wide based on functionally-adjusted Combined Annotation Dependent Depletion (CADD) scores of variants observed in the gnomAD populations, which are referred to as "CADD regions". (2) A region-dependent filtering of rare variants is applied in each CADD region. (3) A functionally-informed burden test is performed with sub-scores computed for each genomic category within each CADD region. Both on simulations and real data, RAVA-FIRST was found to outperform other WGS-based RVAT. Applied to a WGS dataset of venous thromboembolism patients, we identified an intergenic region on chromosome 18 enriched for rare variants in early-onset patients. This region that was missed by standard sliding windows procedures is included in a TAD region that contains a strong candidate gene. RAVA-FIRST enables new investigations of rare non-coding variants in complex diseases, facilitated by its implementation in the R package Ravages.

Ravages: An R package for the simulation and analysis of rare variants in multicategory phenotypes.

Current software packages for the analysis and the simulations of rare variants are only available for binary and continuous traits. Ravages provides solutions in a single R package to perform rare variant association tests for multicategory, binary and continuous phenotypes, to simulate datasets under different scenarios and to compute statistical power. Association tests can be run in the whole genome thanks to C++ implementation of most of the functions, using either RAVA-FIRST, a recently developed strategy to filter and analyse genome-wide rare variants, or user-defined candidate regions. Ravages also includes a simulation module that generates genetic data for cases who can be stratified into several subgroups and for controls. Through comparisons with existing programmes, we show that Ravages complements existing tools and will be useful to study the genetic architecture of complex diseases. Ravages is available on the CRAN at https://cran.r-project.org/web/packages/Ravages/ and maintained on Github at https://github.com/genostats/Ravages.

Do children with mathematical learning disabilities use the inversion principle to solve three-term arithmetic problems?: The impact of presentation mode.

Numerical inversion is the ability to understand that addition is the opposite of subtraction and vice versa. Three-term arithmetic problems can be solved without calculation using this conceptual shortcut. To verify that this principle is used, inverse problems (a + b - b) can be compared with standard problems (a + b - c). If this principle is used, performance on inverse problems will be higher than performance on standard problems because no calculation is required. To our knowledge, this principle has not been previously studied in children with mathematical learning disabilities (MLD). Our objectives were (a) to study whether 10-year-olds with MLD are able to use this conceptual principle in three-term arithmetic problems and (b) to evaluate the impact of the presentation mode. A total of 64 children with or without MLD solved three-term arithmetic problems (inverse and standard) in two presentation modes (symbolic and picture). The results showed that even though children with MLD have difficulties in performing arithmetic problems, they can do so when the inverse problem is presented with pictures. The picture presentation mode allowed children with MLD to efficiently identify and use the conceptual inversion shortcut and thus to achieve a similar performance to that of typically developing children. These results provide interesting perspectives for the care of children with MLD.

Collection and storage of human white blood cells for analysis of DNA damage and repair activity using the comet assay in molecular epidemiology studies.

DNA damage and repair activity are often assessed in blood samples from humans in different types of molecular epidemiology studies. However, it is not always feasible to analyse the samples on the day of collection without any type of storage. For instance, certain studies use repeated sampling of cells from the same subject or samples from different subjects collected at different time-points, and it is desirable to analyse all these samples in the same comet assay experiment. In addition, flawless comet assay analyses on frozen samples open up the possibility of using this technique on biobank material. In this article we discuss the use of cryopreserved peripheral blood mononuclear cells (PBMCs), buffy coat (BC) and whole blood (WB) for analysis of DNA damage and repair using the comet assay. The published literature and the authors' experiences indicate that various types of blood samples can be cryopreserved with only a minor effect on the basal level of DNA damage. There is evidence to suggest that WB and PBMCs can be cryopreserved for several years without much effect on the level of DNA damage. However, care should be taken when cryopreserving WB and BCs. It is possible to use either fresh or frozen samples of blood cells, but results from fresh and frozen cells should not be used in the same dataset. The article outlines detailed protocols for the cryopreservation of PBMCs, BCs and WB samples.

Mixed logistic regression in genome-wide association studies.

BACKGROUND: Mixed linear models (MLM) have been widely used to account for population structure in case-control genome-wide association studies, the status being analyzed as a quantitative phenotype. Chen et al. proved in 2016 that this method is inappropriate in some situations and proposed GMMAT, a score test for the mixed logistic regression (MLR). However, this test does not produces an estimation of the variants' effects. We propose two computationally efficient methods to estimate the variants' effects. Their properties and those of other methods (MLM, logistic regression) are evaluated using both simulated and real genomic data from a recent GWAS in two geographically close population in West Africa. RESULTS: We show that, when the disease prevalence differs between population strata, MLM is inappropriate to analyze binary traits. MLR performs the best in all circumstances. The variants' effects are well evaluated by our methods, with a moderate bias when the effect sizes are large. Additionally, we propose a stratified QQ-plot, enhancing the diagnosis of p values inflation or deflation when population strata are not clearly identified in the sample. CONCLUSION: The two proposed methods are implemented in the R package milorGWAS available on the CRAN. Both methods scale up to at least 10,000 individuals. The same computational strategies could be applied to other models (e.g. mixed Cox model for survival analysis).

Rare variant association testing for multicategory phenotype.

Genetic association studies have provided new insights into the genetic variability of human complex traits with a focus mainly on continuous or binary traits. Methods have been proposed to take into account disease heterogeneity between subgroups of patients when studying common variants but none was specifically designed for rare variants. Because rare variants are expected to have stronger effects and to be more heterogeneously distributed among cases than common ones, subgroup analyses might be particularly attractive in this context. To address this issue, we propose an extension of burden tests by using a multinomial regression model, which enables association tests between rare variants and multicategory phenotypes. We evaluated the type I error and the power of two burden tests, CAST and WSS, by simulating data under different scenarios. In the case of genetic heterogeneity between case subgroups, we showed an advantage of multinomial regression over logistic regression, which considers all the cases against the controls. We replicated these results on real data from Moyamoya disease where the burden tests performed better when cases were stratified according to age-of-onset. We implemented the functions for association tests in the R package "Ravages" available on Github.

First genome-wide association study of non-severe malaria in two birth cohorts in Benin.

Recent research efforts to identify genes involved in malaria susceptibility using genome-wide approaches have focused on severe malaria. Here, we present the first GWAS on non-severe malaria designed to identify genetic variants involved in innate immunity or innate resistance mechanisms. Our study was performed on two cohorts of infants from southern Benin (525 and 250 individuals used as discovery and replication cohorts, respectively) closely followed from birth to 18-24 months of age, with an assessment of a space- and time-dependent environmental risk of exposure. Both the recurrence of mild malaria attacks and the recurrence of malaria infections as a whole (symptomatic and asymptomatic) were considered. Post-GWAS functional analyses were performed using positional, eQTL, and chromatin interaction mapping to identify the genes underlying association signals. Our study highlights a role of PTPRT, a tyrosine phosphatase receptor involved in STAT3 pathway, in the protection against both mild malaria attacks and malaria infections (p = 9.70 × 10-8 and p = 1.78 × 10-7, respectively, in the discovery cohort). Strong statistical support was also found for a role of MYLK4 (meta-analysis, p = 5.29 × 10-8 with malaria attacks), and for several other genes, whose biological functions are relevant in malaria infection. Results shows that GWAS on non-severe malaria can successfully identify new candidate genes and inform physiological mechanisms underlying natural protection against malaria.

Time knowledge acquisition in children aged 6 to 11 years and its relationship with numerical skills.

Acquisition of time knowledge (TK; the correct representation and use of time units) is linked to the development of numerical abilities, but this relationship has not been investigated in children. The current study examined the acquisition of TK and its association with numerical skills. A total of 105 children aged 6 to 11 years were interviewed with our Time Knowledge Questionnaire (TKQ), developed for purposes of this study, and the Zareki-R, a battery for the evaluation of number processing and mental calculation. The TKQ assessed conventional time knowledge (temporal orientation, temporal sequences, relationships between time units, and telling the time on a clock), estimation of longer durations related to birthday and life span, and estimation of the duration of the interview. Time knowledge increased with age, especially from 6 to 8 years, and was strongly linked to numerical skills. Regression analyses showed that four numerical components were implicated in TK: academic knowledge of numbers and number facts (e.g., reading Arabic numerals, mental calculation), number line estimation (e.g., correspondence between a number and a distance), contextual estimation (e.g., many/few leaves on a tree, children in a family), and numerical tasks involving verbal working memory (e.g., comparison of numbers presented orally). Numerical correlations with TK varied according to children's age; subtests based on academic knowledge of numbers, working memory, and number line estimation were linked with TK in the younger children, but only contextual estimation was associated with TK in the older children.

A Bagged, Partially Linear, Tree-Based Regression Procedure for Prediction and Variable Selection.

OBJECTIVES: In genomics, variable selection and prediction accounting for the complex interrelationships between explanatory variables represent major challenges. Tree-based methods are powerful alternatives to classical regression models. We have recently proposed the generalized, partially linear, tree-based regression (GPLTR) procedure that integrates the advantages of generalized linear regression (allowing the incorporation of confounding variables) and of tree-based models. In this work, we use bagging to address a classical concern of tree-based methods: their instability. METHODS: We present a bagged GPLTR procedure and three scores for variable importance. The prediction accuracy and the performance of the scores are assessed by simulation. The use of this procedure is exemplified by the analysis of a lung cancer data set. The aim is to predict the epidermal growth factor receptor (EGFR) mutation based on gene expression measurements, taking into account the ethnicity (confounder variable) and perform variable selection. RESULTS: The procedure performs well in terms of prediction accuracy. The scores differentiate predictive variables from noise variables. Based on a lung adenocarcinoma data set, the procedure achieves good predictive performance for EGFR mutation and selects relevant genes. CONCLUSION: The proposed bagged GPLTR procedure performs well for prediction and variable selection.

The Use of the Linear Mixed Model in Human Genetics.

We give a short but detailed review of the methods used to deal with linear mixed models (restricted likelihood, AIREML algorithm, best linear unbiased predictors, etc.), with a few original points. Then we describe three common applications of the linear mixed model in contemporary human genetics: association testing (pathways analysis or rare variants association tests), genomic heritability estimates, and correction for population stratification in genome-wide association studies. We also consider the performance of best linear unbiased predictors for prediction in this context, through a simulation study for rare variants in a short genomic region, and through a short theoretical development for genome-wide data. For each of these applications, we discuss the relevance and the impact of modeling genetic effects as random effects.

A new scoring system in cancer genetics: application to criteria for BRCA1 and BRCA2 mutation screening.

BACKGROUND: In hereditary forms of cancer due to mutations of genes such as BRCA1 and BRCA2, methods have been proposed to predict the presence of a mutation in a family. METHODS: Relying on carriage probability computation is the most predictive, but scores are a good proxy and avoid using computer software. An empirical method, the Manchester scoring system, has been elaborated for BRCA1 and BRCA2 mutation identification. We propose a general scoring system based on a transformation of the carriage probability. Up to an approximation, the transformed carriage probability becomes an additive score. We applied this new scoring system to the diagnosis of BRCA1-associated and BRCA2-associated breast-ovarian cancer predisposition. Using simulations, its performance was evaluated and compared with that of the Manchester scoring system and of the exact probability. Finally, the score system was used on a sample of 4563 families screened for BRCA1 and BRCA2 mutations. RESULTS: The performance of the new scoring system was superior to the Manchester scoring system, but the probability computation remained the most predictive. The better performance of the new scoring system was attributed to accounting for unaffected family members and for the degree of kinship of relatives with the proband. CONCLUSIONS: The new scoring system has a theoretical basis and may be applied to any cancer family syndrome and, more generally, to any disease with monogenic subentities, in which the causal gene mutations have been identified. It will be easily modified when additional predictive factors are found.

Inbreeding coefficient estimation with dense SNP data: comparison of strategies and application to HapMap III.

BACKGROUND/AIMS: If the parents of an individual are related, it is possible for the individual to have received at 1 locus 2 identical-by-descent alleles that are copies of a single allele carried by the parents' common ancestor. The inbreeding coefficient measures the probability of this event and increases with increasing relatedness between the parents. It is traditionally computed from the observed inbreeding loops in the genealogies and its accuracy thus depends on the depth and reliability of the genealogies. With the availability of genome-wide genetic data, it has become possible to compute a genome-based inbreeding coefficient f, and different methods have been developed to estimate f and identify inbred individuals in a sample from the observed patterns of homozygosity at markers. METHODS: For this paper, we performed simulations with known genealogies using different SNP panels with different levels of linkage disequilibrium (LD) to compare several estimators of f, including single-point estimates, methods based on the length of runs of homozygosity (ROHs) and different methods that use hidden Markov models (HMMs). We also compared the performances of some of these estimators to identify inbred individuals in a sample using either HMM likelihood ratio tests or an adapted version of the ERSA software. RESULTS: Single-point methods were found to have higher standard deviations than other methods. ROHs gave the best estimates provided the correct length threshold is known. HMMs on sparse data gave equivalent or better results than HMMs modeling LD. Provided LD is correctly accounted for, the inbreeding estimates were very similar using the different SNP panels. The HMM likelihood ratio tests were found to perform better at detecting inbred individuals in a sample than the adapted ERSA. All methods accurately detected inbreeding up to second-cousin offspring. We applied the best method on release 3 of the HapMap phase III project, found up to 4% of inbred individuals, and created HAP1067, an unrelated and outbred dataset of this release. CONCLUSIONS: We recommend using HMMs on multiple sparse maps to estimate and detect inbreeding in large samples. If the sample of individuals is too small to estimate allele frequencies, we advise to estimate them on reference panels or to use 1,500-kb ROHs. Finally, we suggest to investigators using HapMap to be careful with inbred individuals, especially in the GIH (Gujarati Indians from Houston in Texas) population.

Using affected sib-pairs to uncover rare disease variants.

OBJECTIVE: We propose a new test for rare variant mapping, based on an affected sib-pair sample and a control sample. In each sib-pair, only the index case needs to be sequenced, and the number of alleles shared identical-by-descent between the sibs is used as complementary information. The test makes use of both association and linkage information. We compare this test to the Armitage test on case-control data, with cases either from the general population of cases or from a sample of cases having an affected sib. METHODS: A score test based on the likelihood in a multiplicative risk model is proposed. Its power is estimated by simulations and compared to Armitage test's power. RESULTS: The affected sib-pairs design allows a tremendous gain of power over the case-control design (from 1 to 99% for a moderate sample size and relative risk values around 3, at an α level of 10(-11)). When cases are ascertained in a sample of cases having an affected sib, the use of linkage information in our test allows a gain of power of more than 20% in certain situations. CONCLUSION: We demonstrate the interest in using familial data and both association and linkage information for rare variant mapping.

Could inbred cases identified in GWAS data succeed in detecting rare recessive variants where affected sib-pairs have failed?

To detect fully penetrant rare recessive variants that could constitute Mendelian subentities of complex diseases, we propose a novel strategy, the HBD-GWAS strategy, which can be applied to genome-wide association study (GWAS) data. This strategy first involves the identification of inbred individuals among cases using the genome-wide SNP data and then focuses on these inbred affected individuals and searches for genomic regions of shared homozygosity by descent that could harbor rare recessive disease-causing variants. In this second step, analogous to homozygosity mapping, a heterogeneity lod-score, HFLOD, is computed to quantify the evidence of linkage provided by the data. In this paper, we evaluate this strategy theoretically under different scenarios and compare its performances with those of linkage analysis using affected sib-pair (ASP) data. If cases affected by these Mendelian subentities are not enriched in the sample of cases, the HBD-GWAS strategy has almost no power to detect them, unless they explain an important part of the disease prevalence. The HBD-GWAS strategy outperforms the ASP linkage strategy only in a very limited number of situations where there exists a strong allelic heterogeneity. When several rare recessive variants within the same gene are involved, the ASP design indeed often fails to detect the gene, whereas, by focusing on inbred individuals using the HBD-GWAS strategy, the gene might be detected provided very large samples of cases are available.

Comorbidity and metabolic context are crucial factors determining neurological sequelae of hypoglycaemia.

AIM: To determine risk factors for neurological sequelae following hypoglycemia. METHOD: We analysed the neurological outcome in 164 patients (mean age 10y 10mo, SD 5.9) following hypoglycemia due to three diseases with various metabolic contexts, different ages at onset, and combinations with comorbidity (fever/infection, hypoxia/ischemia): glycogen storage disease type I (GSDI) (21 patients, mean age at first hypoglycemic episode 3.8mo, SD 3.5); fatty acid ß-oxidation defects (FAOD) (29 patients, mean age at first hypoglycemic episode 14.8mo, SD 12.6); and hyperinsulinism (HIns) (114 patients, mean age at first hypoglycemic episode 2.3mo, SD 4.7). RESULTS: Risk factors of poor neurological outcome were aetiology (p<0.006), comorbidity (p<0.001), and prolonged convulsions (p<0.001). Ordinal logistic regression showed that comorbidity (p<0.001) and status epilepticus (p=0.002) were the main determinants of sequelae. Asymptomatic hypoglycemia did not lead to sequelae, whatever the aetiology. Age was not correlated to sequelae, whatever the aetiology. The highest prevalence of hypoglycemic sequelae was found in FAOD and HIns combined with comorbidity, the lowest in GSDI (p<0.001) in which hypoglycemia is often asymptomatic, associated with increased plasma lactate, and rarely combined with comorbidity. INTERPRETATION: Hypoglycemia is severely deleterious for the brain in the context of fever/infection and/or hypoxia/ischemia, and status epilepticus. The metabolic context providing alternative fuels may improve neurological outcome.

Extension of SKAT to multi-category phenotypes through a geometrical interpretation.

Rare genetic variants are expected to play an important role in disease and several statistical methods have been developed to test for disease association with rare variants, including variance-component tests. These tests however deal only with binary or continuous phenotypes and it is not possible to take advantage of a suspected heterogeneity between subgroups of patients. To address this issue, we extended the popular rare-variant association test SKAT to compare more than two groups of individuals. Simulations under different scenarios were performed that showed gain in power in presence of genetic heterogeneity and minor lack of power in absence of heterogeneity. An application on whole-exome sequencing data from patients with early- or late-onset moyamoya disease also illustrated the advantage of our SKAT extension. Genetic simulations and SKAT extension are implemented in the R package Ravages available on GitHub ( https://github.com/genostats/Ravages ).

DNA damage in circulating leukocytes measured with the comet assay may predict the risk of death.

The comet assay or single cell gel electrophoresis, is the most common method used to measure strand breaks and a variety of other DNA lesions in human populations. To estimate the risk of overall mortality, mortality by cause, and cancer incidence associated to DNA damage, a cohort of 2,403 healthy individuals (25,978 person-years) screened in 16 laboratories using the comet assay between 1996 and 2016 was followed-up. Kaplan-Meier analysis indicated a worse overall survival in the medium and high tertile of DNA damage (p < 0.001). The effect of DNA damage on survival was modelled according to Cox proportional hazard regression model. The adjusted hazard ratio (HR) was 1.42 (1.06-1.90) for overall mortality, and 1.94 (1.04-3.59) for diseases of the circulatory system in subjects with the highest tertile of DNA damage. The findings of this study provide epidemiological evidence encouraging the implementation of the comet assay in preventive strategies for non-communicable diseases.

Adipose Tissue Properties in Tumor-Bearing Breasts.

The tissue stroma plays a major role in tumors' natural history. Most programs for tumor progression are not activated as cell-autonomous processes but under the conditions of cross-talks between tumor and stroma. Adipose tissue is a major component of breast stroma. This study compares adipose tissues in tumor-bearing breasts to those in tumor-free breasts with the intention of defining a signature that could translate into markers of cancer risk. In tumor-bearing breasts, we sampled adipose tissues adjacent to, or distant from the tumor. Parameters studied included: adipocytes size and density, immune cell infiltration, vascularization, secretome and gene expression. Adipose tissues from tumor-bearing breasts, whether adjacent to or distant from the tumor, do not differ from each other by any of these parameters. By contrast, adipose tissues from tumor-bearing breasts have the capacity to secrete twice as much interleukin 8 (IL-8) than those from tumor-free breasts and differentially express a set of 137 genes of which a significant fraction belongs to inflammation, integrin and wnt signaling pathways. These observations show that adipose tissues from tumor-bearing breasts have a distinct physiological status from those from tumor-free breasts. We propose that this constitutive status contributes as a non-cell autonomous process to determine permissiveness for tumor growth.

Validation of Gelbond® high-throughput alkaline and Fpg-modified comet assay using a linear mixed model.

Even if the comet assay has been widely used for decades, there is still a need for controlled studies and good mathematical models to assess the variability of the different versions of this assay and in particular to assess potential intra-experimental variability of the high-throughput comet assay. To address this point, we further validate a high-throughput comet assay that uses hydrophilic polyester film (Gelbond®). Experiments were performed using human peripheral blood mononuclear cells (PBMC) either untreated or treated with different concentration of MMS (methyl methanesulfonate). A positive control for the Fpg (Formamidopyrimidine DNA glycosylase)-modified comet assay (Ro 19-8022 with light) was also included. To quantify the sources of variability of the assay, including intradeposit variability, instead of summarizing DNA damage on 50 cells from a deposit by the mean or median of their percentage DNA tail, we analyzed all logit-transformed data with a linear mixed model. The main source of variation in our experimental data is between cells within the same deposit, suggesting genuine variability in the response of the cells rather than variation caused by technical treatment of cell samples. The second source of variation is the inter-experimental variation (day-to-day experiment); the coefficient of this variation for the control was 13.6%. The variation between deposits in the same experiment is negligible. Moreover, there is no systematic bias because of the position of samples on the Gelbond® film nor the position of the films in the electrophoresis tank. This high-throughput comet assay is thus reliable for various applications. Environ. Mol. Mutagen. 59:595-602, 2018. © 2018 Wiley Periodicals, Inc.

Detecting the dominance component of heritability in isolated and outbred human populations.

Inconsistencies between published estimates of dominance heritability between studies of human genetic isolates and human outbred populations incite investigation into whether such differences result from particular trait architectures or specific population structures. We analyse simulated datasets, characteristic of genetic isolates and of unrelated individuals, before analysing the isolate of Cilento for various commonly studied traits. We show the strengths of using genetic relationship matrices for variance decomposition over identity-by-descent based methods in a population isolate and that heritability estimates in isolates will avoid the downward biases that may occur in studies of samples of unrelated individuals; irrespective of the simulated distribution of causal variants. Yet, we also show that precise estimates of dominance in isolates are demonstrably problematic in the presence of shared environmental effects and such effects should be accounted for. Nevertheless, we demonstrate how studying isolates can help determine the existence or non-existence of dominance for complex traits, and we find strong indications of non-zero dominance for low-density lipoprotein level in Cilento. Finally, we recommend future study designs to analyse trait variance decomposition from ensemble data across multiple population isolates.