Simultaneous SNP selection and adjustment for population structure in high dimensional prediction models

Complex traits are known to be influenced by a combination of environmental factors and rare and common genetic variants. However, detection of such multivariate associations can be compromised by low statistical power and confounding by population structure. Linear mixed effects models (LMM) can account for correlations due to relatedness but have not been applicable in high-dimensional (HD) settings where the number of fixed effect predictors greatly exceeds the number of samples. False positives or false negatives can result from two-stage approaches, where the residuals estimated from a null model adjusted for the subjects' relationship structure are subsequently used as the response in a standard penalized regression model. To overcome these challenges, we develop a general penalized LMM with a single random effect called ggmix for simultaneous SNP selection and adjustment for population structure in high dimensional prediction models. We develop a blockwise coordinate descent algorithm with automatic tuning parameter selection which is highly scalable, computationally efficient and has theoretical guarantees of convergence. Through simulations and three real data examples, we show that ggmix leads to more parsimonious models compared to the two-stage approach or principal component adjustment with better prediction accuracy. Our method performs well even in the presence of highly correlated markers, and when the causal SNPs are included in the kinship matrix. ggmix can be used to construct polygenic risk scores and select instrumental variables in Mendelian randomization studies. Our algorithms are available in an R package available on CRAN (https://cran.r-project.org/package=ggmix).

Gimap3 regulates tissue-specific mitochondrial DNA segregation

Mitochondrial DNA (mtDNA) sequence variants segregate in mutation and tissue-specific manners, but the mechanisms remain unknown. The segregation pattern of pathogenic mtDNA mutations is a major determinant of the onset and severity of disease. Using a heteroplasmic mouse model, we demonstrate that Gimap3, an outer mitochondrial membrane GTPase, is a critical regulator of this process in leukocytes. Gimap3 is important for T cell development and survival, suggesting that leukocyte survival may be a key factor in the genetic regulation of mtDNA sequence variants and in modulating human mitochondrial diseases.

Complexity in the host response to Salmonella Typhimurium infection in AcB and BcA recombinant congenic strains

The host response to Salmonella infection is controlled by its genetic makeup. Using the mouse model of typhoid fever, several genes were found to influence the outcome of Salmonella infection, including Nramp1 (Slc11a1). In order to improve our knowledge of genetic determinants of the mouse response to acute Salmonella Typhimurium infection, we performed a systematic screening of a set of A/J and C57BL/6J recombinant congenic strains (RCS) for their resistance to infection. While we knew that the parental strains differ in their susceptibility to Salmonella because C57BL/6J mice carry a non-functional allele at Nramp1, we hypothesized that other genes would influence the response to Salmonella and segregate in the RCS. We identified several RCS that showed a non-expected phenotype given their known Nramp1 genotype proving that the response to Salmonella in A/J and C57BL/6J mice is complex. Based on these findings, we selected two RCS for generation of fully informative F2 crosses, (AcB61 x 129S6) and (AcB64 x DBA/2J). Genetic analyses performed on these crosses identified five novel Salmonella susceptibility QTL mapping to chromosomes 3 (Ity4), 2 (Ity5), 14 (Ity6), 7 (Ity7) and 15 (Ity8). These results illustrate the genetic complexity associated with the mouse response to Salmonella Typhimurium.

Complex genetic control of susceptibility to Mycobacterium bovis (Bacille Calmette-Guérin) infection in wild-derived Mus spretus mice

Susceptibility to Mycobacterium bovis Bacille Calmette-Guérin (BCG) is genetically controlled by Nramp1 (Slc11a1). Inbred mouse strains harbor either the resistance (Nramp1(G169)) or the susceptibility (Nramp1(D169)) allele at Nramp1. Mus spretus (Nramp1(G169); SPRET/EiJ) is shown to display an intermediate level of BCG replication in the spleen (log(10) colony-forming units (CFU) approximately 5), compared to resistant A/J (log(10)CFU approximately 4.0) and susceptible C57BL/6J (log(10)CFU approximately 6.0) mice. The presence of genetic modifiers of Nramp1-dependent susceptibility to M. bovis (BCG) infection in Mus spretus was analyzed by whole-genome scanning in 175 mice of an informative (C57BL/6J x SPRET/EiJ) x C57BL/6J backcross. Nramp1 showed a major effect (D1Mcg4, P<1e(-4)), but additional single marker effects were identified on chromosomes 4 (D4Mit150) and x (DXMit249) in male mice, and on chromosome 9 (D9Mit77) and 17 (D17Mit81) in female mice. A strong interaction between Nramp1 and the major histocompatibility locus was also noted in female mice. The mapped loci may act as modifiers of Nramp1 action, and constitute novel entry points for the parallel search of loci regulating susceptibility to mycobacterial infections in humans.

Common polymorphisms in the promoter of the visfatin gene (PBEF1) influence plasma insulin levels in a French-Canadian population

The adipokine visfatin (PBEF1) exhibits insulin-mimetic effects and correlates strongly with visceral adiposity. We sequenced visfatin gene exons and 1,480 bp of the promoter in 23 individuals, including 18 individuals from the Quebec Family Study (QFS) with varying degrees of abdominal visceral fat, assessed by computed tomography, and 5 individuals from the Saguenay-Lac-Saint-Jean region of Québec. We identified a synonymous polymorphism in exon 7 (SER301SER) but no nonsynonymous mutations. We observed an additional 10 polymorphisms, including 5 intronic, 4 within the promoter, and 1 within the 3' untranslated region. Further promoter sequencing (816 bp) identified five additional single nucleotide polymorphisms (SNPs) in the QFS population. To investigate the role of visfatin gene variants in obesity-related phenotypes, we genotyped a total of 13 SNPs in the promoter region of the gene. From these, we analyzed the seven common SNPs in the QFS sample (918 participants from 208 families). A significant association was found between two SNPs (rs9770242 and rs1319501), in perfect linkage disequilibrium, and fasting insulin levels (P = 0.002). These SNPs were also associated with fasting glucose (P <or= 0.02). In addition, a more distal SNP (rs7789066) was significantly associated with the apolipoprotein B component of VLDL (P = 0.012).

SLC34A3 mutations in patients with hereditary hypophosphatemic rickets with hypercalciuria predict a key role for the sodium-phosphate cotransporter NaPi-IIc in maintaining phosphate homeostasis

Hereditary hypophosphatemic rickets with hypercalciuria (HHRH) is a rare disorder of autosomal recessive inheritance that was first described in a large consanguineous Bedouin kindred. HHRH is characterized by the presence of hypophosphatemia secondary to renal phosphate wasting, radiographic and/or histological evidence of rickets, limb deformities, muscle weakness, and bone pain. HHRH is distinct from other forms of hypophosphatemic rickets in that affected individuals present with hypercalciuria due to increased serum 1,25-dihydroxyvitamin D levels and increased intestinal calcium absorption. We performed a genomewide linkage scan combined with homozygosity mapping, using genomic DNA from a large consanguineous Bedouin kindred that included 10 patients who received the diagnosis of HHRH. The disease mapped to a 1.6-Mbp region on chromosome 9q34, which contains SLC34A3, the gene encoding the renal sodium-phosphate cotransporter NaP(i)-IIc. Nucleotide sequence analysis revealed a homozygous single-nucleotide deletion (c.228delC) in this candidate gene in all individuals affected by HHRH. This mutation is predicted to truncate the NaP(i)-IIc protein in the first membrane-spanning domain and thus likely results in a complete loss of function of this protein in individuals homozygous for c.228delC. In addition, compound heterozygous missense and deletion mutations were found in three additional unrelated HHRH kindreds, which supports the conclusion that this disease is caused by SLC34A3 mutations affecting both alleles. Individuals of the investigated kindreds who were heterozygous for a SLC34A3 mutation frequently showed hypercalciuria, often in association with mild hypophosphatemia and/or elevations in 1,25-dihydroxyvitamin D levels. We conclude that NaP(i)-IIc has a key role in the regulation of phosphate homeostasis.

Segregation of urine calcium excretion in families ascertained for nephrolithiasis: evidence for a major gene

BACKGROUND

The quantitative genetics of urine calcium excretion has not been established. It is a trait of interest because hypercalciuria is commonly found in subjects with nephrolithiasis. The aim of this study was to model the segregation of this trait in a sample of French-Canadian families ascertained through a stone former.

METHODS

Major gene, polygenic, and mixed models were fit to 24-hour urine calcium excretion from 567 individuals in 221 nuclear families, while simultaneously taking into account gender, age at examination, body mass index (BMI), and the use of thiazide drugs. The nuclear families were extracted from 154 pedigrees, some of which were four generations, with at least two siblings with a history of calcium stones.

RESULTS

All the proposed genetic models fit the data significantly better than the null model. The most parsimonious model was the mixed codominant/polygenic model but it was statistically indistinguishable from the single-gene codominant model. In both of these models the heritability attributable to the major gene was estimated to be 0.58.

CONCLUSION

Our results suggest that a major gene with a relatively large effect on variation in urine calcium excretion is segregating in French-Canadian families with stone formers. This implies that the power of quantitative trait segregation analysis of urine calcium excretion may be increased in these families, and results indicate that it should be feasible to genetically map the quantitative trait locus.

Localisation of a gene for mucopolysaccharidosis IIIC to the pericentromeric region of chromosome 8

Mucopolysaccharidosis type IIIC (MPS IIIC, or Sanfilippo syndrome C) is a rare lysosomal storage disorder caused by a deficiency of acetyl-coenzyme A:alpha-glucosaminide-N-acetyltransferase. Patients develop progressive neuropsychiatric problems, mental retardation, hearing loss, and relatively minor visceral manifestations. The pattern of transmission is consistent with an autosomal recessive mode of inheritance. The aim of this study was to find a locus for MPS IIIC using a homozygosity mapping approach. A genomewide scan was performed on DNA from 27 affected individuals and 17 of their unaffected relatives. Additional patients were recruited, and DNA was obtained from a total of 44 affected individuals and 18 unaffected family members from 31 families from 10 countries. A working candidate interval was defined by looking for excess homozygosity in patients compared with their relatives. Additional markers were genotyped in regions of interest. Linkage analysis was performed to support the informal analysis. Inspection of the genomewide scan data showed apparent excess homozygosity in patients compared with their relatives for markers on chromosome 8. Additional genotyping identified 15 consecutive markers (from D8S1051 to D8S2332) in an 8.3 cM interval for which the genotypes of affected siblings were identical in state. A maximum multipoint lod score of 10.61 was found at marker D8S519. A locus for MPS IIIC maps to an 8.3 cM (16 Mbp) interval in the pericentromeric region of chromosome 8.

Mapping of interactions and mouse congenic strains identified novel epistatic QTLs controlling the persistence of Salmonella Enteritidis in mice

The host response to infection in humans is multifactorial and involves the complex interaction between two genomes (the host and the pathogen) and the environment. Using an experimental mouse model of chronic infection, we have previously identified the individual effect of three significant and one suggestive quantitative trait loci (QTLs) (Ses1, Ses2, Ses3 and Ses1.1) on Salmonella Enteritidis persistence in target organs of 129S6/SvEvTac mice. Congenic strain construction was performed by transferring each of these QTLs from C57BL/6J onto the 129S6/SvEvTac background, and phenotypic analysis confirmed that Ses1 and Ses1.1 contribute to bacterial clearance. Additional QTLs regulating Salmonella carriage in 129S6/SvEvTac mice were identified using a two-locus epistasis QTL linkage mapping approach conducted separately in females and males. The epistatic model for females included the individual effect of Ses3 and two significant interactions (Ses1-D7Mit267 and Ses1-DXMit48) accounting for 47% of the total phenotypic variance. The model for males included the individual effect of Ses1.1, three interactions (Ses1-D9Mit218, D2Mit197-D4Mit2 and D3Mit256-D13Mit36) and explained 47% of the phenotypic variance. Our results suggest that the oligogenic nature of Salmonella persistence and epistasis are important constituents of the genetic architecture of the host response to chronic Salmonella infection.

Bowen-Conradi syndrome: a clinical and genetic study

The purpose of the study was to delineate the anomalies and the natural life history of persons with the Bowen-Conradi syndrome [Bowen and Conradi 1976: Birth Defects: Orig Artic Ser XII(6):101-108]. We ascertained 39 cases and personally examined almost all. For those who were not seen, their clinical record were scrutinized. Pedigree analysis of all 39 was done and kinship coefficients computed. The birth prevalence was estimated to be 1/355 live births.

Allelic variation in TLR4 is linked to susceptibility to Salmonella enterica serovar Typhimurium infection in chickens

Toll-like receptor 4 (TLR4) is part of a group of evolutionarily conserved pattern recognition receptors involved in the activation of the immune system in response to various pathogens and in the innate defense against infection. We describe here the cloning and characterization of the avian orthologue of mammalian TLR4. Chicken TLR4 encodes a 843-amino-acid protein that contains a leucine-rich repeat extracellular domain, a short transmembrane domain typical of type I transmembrane proteins, and a Toll-interleukin-1R signaling domain characteristic of all TLR proteins. The chicken TLR4 protein shows 46% identity (64% similarity) to human TLR4 and 41% similarity to other TLR family members. Northern blot analysis reveals that TLR4 is expressed at approximately the same level in all tissues tested, including brain, thymus, kidney, intestine, muscle, liver, lung, bursa of Fabricius, heart, and spleen. The probe detected only one transcript of ca. 4.4 kb in length for all tissues except muscle where the size of TLR4 mRNA was ca. 9.6 kb. We have mapped TLR4 to microchromosome E41W17 in a region harboring the gene for tenascin C and known to be well conserved between the chicken and mammalian genomes. This region of the chicken genome was shown previously to harbor a Salmonella susceptibility locus. By using linkage analysis, TLR4 was shown to be linked to resistance to infection with Salmonella enterica serovar Typhimurium in chickens (likelihood ratio test of 10.2, P = 0.00138), suggesting a role of TLR4 in the host response of chickens to Salmonella infection.

Nuclear genetic control of mitochondrial DNA segregation

Mammalian mitochondrial DNA (mtDNA) is a high copy-number, maternally inherited genome that codes for a small number of essential proteins involved in oxidative phosphorylation. Mutations in mtDNA are responsible for a broad spectrum of clinical disorders. The segregation pattern of pathogenic mtDNA mutants is an important determinant of the nature and severity of mitochondrial disease, but it varies with the specific mutation, cell type and nuclear background and generally does not correlate well with mitochondrial dysfunction. To identify nuclear genes that modify the segregation behavior of mtDNA, we used a heteroplasmic mouse model derived from two inbred strains (BALB/c and NZB; ref. 12), in which we had previously demonstrated tissue-specific and age-dependent directional selection for different mtDNA genotypes in the same mouse. Here we show that this phenotype segregates in F2 mice from a genetic cross (BALB/c x CAST/Ei) and that it maps to at least three quantitative-trait loci (QTLs). Genome-wide scans showed linkage of the trait to loci on Chromosomes 2, 5 and 6, accounting for 16-35% of the variance in the trait, depending on the tissue and age of the mouse. This is the first genetic evidence for nuclear control of mammalian mtDNA segregation.

Identification of genetic loci controlling bacterial clearance in experimental Salmonella enteritidis infection: an unexpected role of Nramp1 (Slc11a1) in the persistence of infection in mice

The Gram-negative bacteria, Salmonella, cause a broad spectrum of clinical diseases in both animals and humans ranging from asymptomatic carriage to life-threatening sepsis. We have developed a model to study the contribution of genetic factors to the susceptibility of 129sv and C57BL/6J inbred mice to Salmonella enteritidis during the late phase of infection. C57BL/6J mice were able to eliminate completely sublethal inoculums of S. enteritidis from their reticuloendothelial system, whereas 129sv mice could not even after 60 days post inoculation. A genome scan performed on 302 (C57BL/6J x 129sv) F2 progeny identified three dominant loci (designated Ses1 to Ses3) that are associated with disease susceptibility in 129sv mice. Two highly significant linkages were identified on chromosomes 1 (Ses1) and 7 (Ses2) with respective LOD scores of 9.9 (P = 1.4 x 10(-11)) at D1Mcg5 and 4.0 (P = 1.9 x 10(-5)) at D7Mit62. One highly suggestive QTL was located on chromosomes15 (Ses3) with a LOD score 3.4 (P = 1.2 x 10(-4)). The estimated effects of Ses1, Ses2 and Ses3 on the bacterial clearance were greater in females. Using a model of three loci, with interaction between Ses1 and Ses2 and sex as a covariate, the three QTLs explained 32% of the phenotypic variance. The candidacy of Nramp1 as the gene for Ses1 was evaluated using mice carrying a null allele at Nramp1 (129sv-Nramp1(tm1Mcg)). These mice have a significantly lower spleen bacterial load compared to the wild-type 129sv mice, strongly suggesting the involvement of Nramp1 in controlling S. enteritidis clearance during the late phase of infection.

5' flanking variants of resistin are associated with obesity

Diabetes and obesity have long been known to be related. The recently characterized adipocyte hormone resistin (also called FIZZ3/ADSF) has been implicated as a molecular link between impaired glucose tolerance (IGT) and obesity in mice. A search for sequence variants at the human resistin locus identified nine single-nucleotide polymorphisms (SNPs) but no coding variants. An investigation into the association of these SNPs with diabetes and obesity revealed two 5' flanking variants (g.-537 and g.-420), in strong linkage disequilibrium, that are associated with BMI. In nondiabetic individuals from the Quebec City area and the Saguenay-Lac-St-Jean region of Quebec, the g.-537 mutation (allelic frequency = 0.04) was significantly associated with an increase in BMI (P = 0.03 and P = 0.01, respectively). When the data from these two populations were combined and adjusted for age and sex, both the g.-537 (odds ratio [OR] 2.72, 95% CI 1.28-5.81) and the g.-420 variants (1.58, 1.06-2.35) were associated with an increased risk for a BMI > or =30 kg/m(2). In contrast, in case/control and family-based study populations from Scandinavia, we saw no effect on BMI with either of these promoter variants. No association was seen with diabetes in any of the population samples.

Pedigree selection and tests of linkage in a Hutterite asthma pedigree

We explored methods for kinship and linkage analysis in a Hutterite pedigree comprising 1,544 individuals, 72 of whom were diagnosed with asthma. Subpedigrees were selected by (a) identifying nuclear families containing asthmatics, (b) identifying couples with many asthmatic descendants in an ad hoc manner, and (c) finding the most recent common ancestors of all asthmatics. Markov chain Monte Carlo (MCMC) methods were used to estimate allele sharing in the larger subpedigrees and transmission/disequilibrium tests were performed on nuclear families. On chromosome 5q near the cytokine cluster, modest evidence for linkage to asthma was obtained. Using MCMC, we were able to evaluate the evidence for linkage in complex subpedigrees of several hundred individuals, and hence, incorporate some of the co-ancestry of this founder population.