A Missense Variant in PTPN22 is a Risk Factor for Drug-induced Liver Injury.

BACKGROUND & AIMS: We performed genetic analyses of a multiethnic cohort of patients with idiosyncratic drug-induced liver injury (DILI) to identify variants associated with susceptibility. METHODS: We performed a genome-wide association study of 2048 individuals with DILI (cases) and 12,429 individuals without (controls). Our analysis included subjects of European (1806 cases and 10,397 controls), African American (133 cases and 1,314 controls), and Hispanic (109 cases and 718 controls) ancestry. We analyzed DNA from 113 Icelandic cases and 239,304 controls to validate our findings. RESULTS: We associated idiosyncratic DILI with rs2476601, a nonsynonymous polymorphism that encodes a substitution of tryptophan with arginine in the protein tyrosine phosphatase, nonreceptor type 22 gene (PTPN22) (odds ratio [OR] 1.44; 95% confidence interval [CI] 1.28-1.62; P = 1.2 × 10-9 and replicated the finding in the validation set (OR 1.48; 95% CI 1.09-1.99; P = .01). The minor allele frequency showed the same effect size (OR > 1) among ethnic groups. The strongest association was with amoxicillin and clavulanate-associated DILI in persons of European ancestry (OR 1.62; 95% CI 1.32-1.98; P = 4.0 × 10-6; allele frequency = 13.3%), but the polymorphism was associated with DILI of other causes (OR 1.37; 95% CI 1.21-1.56; P = 1.5 × 10-6; allele frequency = 11.5%). Among amoxicillin- and clavulanate-associated cases of European ancestry, rs2476601 doubled the risk for DILI among those with the HLA risk alleles A*02:01 and DRB1*15:01. CONCLUSIONS: In a genome-wide association study, we identified rs2476601 in PTPN22 as a non-HLA variant that associates with risk of liver injury caused by multiple drugs and validated our finding in a separate cohort. This variant has been associated with increased risk of autoimmune diseases, providing support for the concept that alterations in immune regulation contribute to idiosyncratic DILI.

Minocycline hepatotoxicity: Clinical characterization and identification of HLA-B∗35:02 as a risk factor.

BACKGROUND & AIMS: Minocycline hepatotoxicity can present with prominent autoimmune features in previously healthy individuals. The aim of this study was to identify genetic determinants of minocycline drug-induced liver injury (DILI) in a well-phenotyped cohort of patients. METHODS: Caucasian patients with minocycline DILI underwent genome-wide genotyping and were compared to unexposed population controls. Human leukocyte antigen (HLA) binding of minocycline was assessed using AutoDock Vina. RESULTS: Among the 25 cases, 80% were female, median age was 19years and median latency from drug start to DILI onset was 318days. At presentation, 76% had acute hepatocellular liver injury, median ALT 1,077U/L (range: 63 to 2,333), median bilirubin 4.5mg/dl (range: 0.2 to 16.7), and 90% had a +ANA. During follow-up, 50% were treated with corticosteroids and no participants died or required a liver transplant. A significant association was noted between HLA-B∗35:02 and risk for minocycline DILI; a 16% carrier frequency in DILI cases compared to 0.6% in population controls (odds ratio: 29.6, 95% CI: 7.8-89.8, p=2.5×10-8). Verification of HLA-B∗35:02 imputation was confirmed by sequence-based HLA typing. HLA-B∗35:02 carriers had similar presenting features and outcomes compared to non-carriers. In silico modeling studies support the hypothesis that direct binding of minocycline to this novel HLA risk allele might be an important initiating event in minocycline DILI. CONCLUSION: HLA-B∗35:02 is a rare HLA allele that was more frequently identified in the 25 minocycline DILI cases compared to population controls. If confirmed in other cohorts, this HLA allele may prove to be a useful diagnostic marker of minocycline DILI. LAY SUMMARY: Development of liver injury following prolonged use of minocycline for acne is a rare but potentially severe form of drug-induced liver injury. Our study demonstrates that individuals who are HLA-B∗35:02 carriers are at increased risk of developing minocycline related liver injury. These results may help doctors more rapidly and confidently diagnose affected patients and possibly reduce the risk of liver injury in individuals receiving minocycline going forward.

Features of Autoimmune Hepatitis in Patients With Drug-induced Liver Injury.

BACKGROUND & AIMS: Drug-induced liver injury (DILI) has features similar to those of other liver diseases including autoimmune hepatitis (AIH). We aimed to characterize the clinical and autoimmune features of liver injury caused by nitrofurantoin, minocycline, methyldopa, or hydralazine. METHODS: We analyzed data from 88 cases of DILI attributed to nitrofurantoin, minocycline, methyldopa, or hydralazine included in the Drug-Induced Liver Injury Network prospective study from 2004 through 2014. Sera were collected from patients at baseline and follow-up examination and tested for levels of immunoglobulin G (IgG), antibodies to nuclear antigen (ANA), smooth muscle (SMA), and soluble liver antigen (SLA). An autoimmune score was derived on the basis of increases in levels of IgG, ANA, SMA, and SLA (assigned values of 0, 1+, or 2+). AIH-associated HLA-DRB1*03:01 and HLA-DRB1*04:01 allele frequencies were compared with those of the general population (controls). RESULTS: Of the 88 cases, 80 were women (91%), 74% had hepatocellular injury, and 25% had severe injury. At the onset of DILI, 39% of cases had increased levels of IgG, 72% had increased levels of ANA, 60% had increased levels of SMA, and none had increases in SLA. A phenotype of autoimmunity (autoimmune score ≥2) was observed in 82% of cases attributed to nitrofurantoin and 73% of cases attributed to minocycline (73%) but only 55% of cases attributed to methyldopa and 43% of cases attributed to hydralazine (P = .16 for nitrofurantoin and minocycline vs methyldopa and hydralazine). We observed a decrease in numbers of serum samples positive for ANA (P = .01) or SMA (P < .001) and in autoimmune scores (P < .001) between DILI onset and follow-up. Similar percentages of patients with DILI had HLA-DRB1*03:01 (15%) and HLA-DRB1*04:01 (9%) as controls (12% and 9%, respectively). CONCLUSIONS: In analysis of data from the DILIN prospective study, we found that most cases of DILI attributed to nitrofurantoin or minocycline and about half of cases that were due to methyldopa and hydralazine have a phenotype of autoimmunity similar to AIH. These features decrease with recovery of the injury and are not associated with the typical HLA alleles found in patients with idiopathic AIH.

An exploratory genome-wide analysis of genetic risk for alcoholic hepatitis.

OBJECTIVES: To elucidate the genetic variability between heavy drinkers with and without alcoholic hepatitis (AH). MATERIALS AND METHODS: An exploratory genome-wide association study (GWAS; NCT02172898) was conducted comparing 90 AH cases with 93 heavy drinking matched controls without liver disease in order to identify variants or genes associated with risk for AH. Individuals were genotyped using the multi-ethnic genotyping array, after which the data underwent conventional quality control. Using bioinformatics tools, pathways associated with AH were explored on the basis of individual variants, and based on genes with a higher 'burden' of functional variation. RESULTS: Although no single variant reached genome-wide significance, an association signal was observed for PNPLA3 rs738409 (p = .01, OR 1.9, 95% CI 1.1-3.1), a common single nucleotide polymorphism that has been associated with a variety of liver-related pathologies including alcoholic cirrhosis. Using the improved gene set enrichment analysis for GWAS tool, it was shown that, based on the single variants' trait-association p-values, multiple pathways were associated with risk for AH with high confidence (false discovery rate [FDR] < 0.05), including several pathways involved in lymphocyte activation and chemokine signaling, which coincides with findings from other research groups. Several Tox Functions and Canonical Pathways were highlighted using Ingenuity Pathway Analysis, with an especially conspicuous role for pathways related to ethanol degradation, which is not surprising considering the phenotype of the genotyped individuals. CONCLUSION: This preliminary analysis suggests a role for PNPLA3 variation and several gene sets/pathways that may influence risk for AH among heavy drinkers.

Leveraging prior information to detect causal variants via multi-variant regression.

Although many methods are available to test sequence variants for association with complex diseases and traits, methods that specifically seek to identify causal variants are less developed. Here we develop and evaluate a Bayesian hierarchical regression method that incorporates prior information on the likelihood of variant causality through weighting of variant effects. By simulation studies using both simulated and real sequence variants, we compared a standard single variant test for analyzing variant-disease association with the proposed method using different weighting schemes. We found that by leveraging linkage disequilibrium of variants with known GWAS signals and sequence conservation (phastCons), the proposed method provides a powerful approach for detecting causal variants while controlling false positives.

The SWine IMputation (SWIM) haplotype reference panel enables nucleotide resolution genetic mapping in pigs.

Genetic mapping to identify genes and alleles associated with or causing economically important quantitative trait variation in livestock animals such as pigs is a major goal in animal genetic improvement. Despite recent advances in high-throughput genotyping technologies, the resolution of genetic mapping in pigs remains poor due in part to the low density of genotyped variant sites. In this study, we overcame this limitation by developing a reference haplotype panel for pigs based on 2259 whole genome-sequenced animals representing 44 pig breeds. We evaluated software combinations and breed composition to optimize the imputation procedure and achieved an average concordance rate in excess of 96%, a non-reference concordance rate of 88%, and an r2 of 0.85. We demonstrated in two case studies that genotype imputation using this resource can dramatically improve the resolution of genetic mapping. A public web server has been developed to allow the pig genetics community to fully utilize this resource. We expect this resource to facilitate genetic mapping and accelerate genetic improvement in pigs.

Genome-wide identification and initial characterization of bovine long non-coding RNAs from EST data.

It has become increasingly clear that the mammalian genomes produce many long non-coding RNAs (lncRNAs). Accumulating evidence suggests important functions for lncRNAs in a variety of biological processes. However, little is known about lncRNA identity and characteristics in cattle. Using public bovine-specific expressed sequence tags sequences, we reconstructed transcript assemblies, from which reference sequences were obtained for RNAs. Intergenic regions with evidence of transcription were screened for putative lncRNAs using the combination of a gene-finding program and a support vector machine-based tool for the calculation of protein-coding potential. A total of 449 putative lncRNAs located in 405 intergenic regions were identified. Characterization of these putative bovine lncRNAs suggests that they are generally expressed in a tissue-specific manner, their GC contents are higher than randomly selected intergenic sequences but are lower than protein-coding genes, and they are moderately conserved among mammals. This is the first genome-wide catalogue of putative intergenic lncRNAs in cattle and provides important targets for functional studies.

Marker-assisted prediction of non-additive genetic values.

It has become increasingly clear from systems biology arguments that interaction and non-linearity play an important role in genetic regulation of phenotypic variation for complex traits. Marker-assisted prediction of genetic values assuming additive gene action has been widely investigated because of its relevance in artificial selection. On the other hand, it has been less well-studied when non-additive effects hold. Here, we explored a nonparametric model, radial basis function (RBF) regression, for predicting quantitative traits under different gene action modes (additivity, dominance and epistasis). Using simulation, it was found that RBF had better ability (higher predictive correlations and lower predictive mean square errors) of predicting merit of individuals in future generations in the presence of non-additive effects than a linear additive model, the Bayesian Lasso. This was true for populations undergoing either directional or random selection over several generations. Under additive gene action, RBF was slightly worse than the Bayesian Lasso. While prediction of genetic values under additive gene action is well handled by a variety of parametric models, nonparametric RBF regression is a useful counterpart for dealing with situations where non-additive gene action is suspected, and it is robust irrespective of mode of gene action.

Application of support vector regression to genome-assisted prediction of quantitative traits.

A byproduct of genome-wide association studies is the possibility of carrying out genome-enabled prediction of disease risk or of quantitative traits. This study is concerned with predicting two quantitative traits, milk yield in dairy cattle and grain yield in wheat, using dense molecular markers as predictors. Two support vector regression (SVR) models, ε-SVR and least-squares SVR, were explored and compared to a widely applied linear regression model, the Bayesian Lasso, the latter assuming additive marker effects. Predictive performance was measured using predictive correlation and mean squared error of prediction. Depending on the kernel function chosen, SVR can model either linear or nonlinear relationships between phenotypes and marker genotypes. For milk yield, where phenotypes were estimated breeding values of bulls (a linear combination of the data), SVR with a Gaussian radial basis function (RBF) kernel had a slightly better performance than with a linear kernel, and was similar to the Bayesian Lasso. For the wheat data, where phenotype was raw grain yield, the RBF kernel provided clear advantages over the linear kernel, e.g., a 17.5% increase in correlation when using the ε-SVR. SVR with a RBF kernel also compared favorably to the Bayesian Lasso in this case. It is concluded that a nonlinear RBF kernel may be an optimal choice for SVR, especially when phenotypes to be predicted have a nonlinear dependency on genotypes, as it might have been the case in the wheat data.

Radial basis function regression methods for predicting quantitative traits using SNP markers.

A challenge when predicting total genetic values for complex quantitative traits is that an unknown number of quantitative trait loci may affect phenotypes via cryptic interactions. If markers are available, assuming that their effects on phenotypes are additive may lead to poor predictive ability. Non-parametric radial basis function (RBF) regression, which does not assume a particular form of the genotype-phenotype relationship, was investigated here by simulation and analysis of body weight and food conversion rate data in broilers. The simulation included a toy example in which an arbitrary non-linear genotype-phenotype relationship was assumed, and five different scenarios representing different broad sense heritability levels (0.1, 0.25, 0.5, 0.75 and 0.9) were created. In addition, a whole genome simulation was carried out, in which three different gene action modes (pure additive, additive+dominance and pure epistasis) were considered. In all analyses, a training set was used to fit the model and a testing set was used to evaluate predictive performance. The latter was measured by correlation and predictive mean-squared error (PMSE) on the testing data. For comparison, a linear additive model known as Bayes A was used as benchmark. Two RBF models with single nucleotide polymorphism (SNP)-specific (RBF I) and common (RBF II) weights were examined. Results indicated that, in the presence of complex genotype-phenotype relationships (i.e. non-linearity and non-additivity), RBF outperformed Bayes A in predicting total genetic values using SNP markers. Extension of Bayes A to include all additive, dominance and epistatic effects could improve its prediction accuracy. RBF I was generally better than RBF II, and was able to identify relevant SNPs in the toy example.

Influence of Genetic Interactions on Polygenic Prediction.

Prediction of phenotypes from genotypes is an important objective to fulfill the promises of genomics, precision medicine and agriculture. Although it's now possible to account for the majority of genetic variation through model fitting, prediction of phenotypes remains a challenge, especially across populations that have diverged in the past. In this study, we designed simulation experiments to specifically investigate the role of genetic interactions in failure of polygenic prediction. We found that non-additive genetic interactions can significantly reduce the accuracy of polygenic prediction. Our study demonstrated the importance of considering genetic interactions in genetic prediction.

Nonparametric methods for incorporating genomic information into genetic evaluations: an application to mortality in broilers.

Four approaches using single-nucleotide polymorphism (SNP) information (F(infinity)-metric model, kernel regression, reproducing kernel Hilbert spaces (RKHS) regression, and a Bayesian regression) were compared with a standard procedure of genetic evaluation (E-BLUP) of sires using mortality rates in broilers as a response variable, working in a Bayesian framework. Late mortality (14-42 days of age) records on 12,167 progeny of 200 sires were precorrected for fixed and random (nongenetic) effects used in the model for genetic evaluation and for the mate effect. The average of the corrected records was computed for each sire. Twenty-four SNPs seemingly associated with late mortality were included in three methods used for genomic assisted evaluations. One thousand SNPs were included in the Bayesian regression, to account for markers along the whole genome. The posterior mean of heritability of mortality was 0.02 in the E-BLUP approach, suggesting that genetic evaluation could be improved if suitable molecular markers were available. Estimates of posterior means and standard deviations of the residual variance were 24.38 (3.88), 29.97 (3.22), 17.07 (3.02), and 20.74 (2.87) for E-BLUP, the linear model on SNPs, RKHS regression, and the Bayesian regression, respectively, suggesting that RKHS accounted for more variance in the data. The two nonparametric methods (kernel and RKHS regression) fitted the data better, having a lower residual sum of squares. Predictive ability, assessed by cross-validation, indicated advantages of the RKHS approach, where accuracy was increased from 25 to 150%, relative to other methods.

Comparison of classification methods for detecting associations between SNPs and chick mortality.

Multi-category classification methods were used to detect SNP-mortality associations in broilers. The objective was to select a subset of whole genome SNPs associated with chick mortality. This was done by categorizing mortality rates and using a filter-wrapper feature selection procedure in each of the classification methods evaluated. Different numbers of categories (2, 3, 4, 5 and 10) and three classification algorithms (naïve Bayes classifiers, Bayesian networks and neural networks) were compared, using early and late chick mortality rates in low and high hygiene environments. Evaluation of SNPs selected by each classification method was done by predicted residual sum of squares and a significance test-related metric. A naïve Bayes classifier, coupled with discretization into two or three categories generated the SNP subset with greatest predictive ability. Further, an alternative categorization scheme, which used only two extreme portions of the empirical distribution of mortality rates, was considered. This scheme selected SNPs with greater predictive ability than those chosen by the methods described previously. Use of extreme samples seems to enhance the ability of feature selection procedures to select influential SNPs in genetic association studies.

Sensitivity to hepatotoxicity due to epigallocatechin gallate is affected by genetic background in diversity outbred mice.

Consumer use of herbal and dietary supplements has recently grown in the United States and, with increased use, reports of rare adverse reactions have emerged. One such supplement is green tea extract, containing the polyphenol epigallocatechin gallate (EGCG), which has been shown to be hepatotoxic at high doses in animal models. The Drug-Induced Liver Injury Network has identified multiple patients who have experienced liver injury ascribed to green tea extract consumption and the relationship to dose has not been straightforward, indicating that differences in sensitivity may contribute to the adverse response in susceptible people. The Diversity Outbred (DO), a genetically heterogeneous mouse population, provides a potential platform for study of interindividual toxicity responses to green tea extract. Within the DO population, an equal exposure to EGCG (50 mg/kg; daily for three days) was found to be tolerated in the majority of mice; however, a small fraction of the animals (16%; 43/272) exhibited severe hepatotoxicity (10-86.8% liver necrosis) that is analogous to the clinical cases. The data indicate that the DO mice may provide a platform for informing risk of rare, adverse reactions that may occur in consumer populations upon ingestion of concentrated herbal products.

Penguin tissue as a proxy for relative krill abundance in East Antarctica during the Holocene.

Antarctic krill (Euphausia superba) is a key component of the Southern Ocean food web. It supports a large number of upper trophic-level predators, and is also a major fishery resource. Understanding changes in krill abundance has long been a priority for research and conservation in the Southern Ocean. In this study, we performed stable isotope analyses on ancient Adélie penguin tissues and inferred relative krill abundance during the Holocene epoch from paleodiets of Adélie penguin (Pygoscelis adeliae), using inverse of δ¹5N (ratio of ¹5N/¹4N) value as a proxy. We find that variations in krill abundance during the Holocene are in accord with episodes of regional climate changes, showing greater krill abundance in cold periods. Moreover, the low δ¹5N values found in modern Adélie penguins indicate relatively high krill availability, which supports the hypothesis of krill surplus in modern ages due to recent hunt for krill-eating seals and whales by humans.

Distributions and impact factors of antimony in topsoils and moss in Ny-Ålesund, Arctic.

The distribution of antimony (Sb) in topsoil and moss (Dicranum angustum) in disturbed and undisturbed areas, as well as coal and gangue, in Ny-Ålesund, Arctic was examined. Results show that the weathering of coal bed could not contribute to the increase of Sb concentrations in topsoil and moss in the study area. The distribution of Sb is partially associated with traffic and historical mining activities. The occurrence of the maximum Sb concentration is due to the contribution of human activities. In addition, the decrease of Sb content in topsoil near the coastline may be caused by the washing of seawater. Compared with topsoils, moss could be a useful tool for monitoring Sb in both highly and lightly polluted areas.

Predicting genetic values: a kernel-based best linear unbiased prediction with genomic data.

Genomic data provide a valuable source of information for modeling covariance structures, allowing a more accurate prediction of total genetic values (GVs). We apply the kriging concept, originally developed in the geostatistical context for predictions in the low-dimensional space, to the high-dimensional space spanned by genomic single nucleotide polymorphism (SNP) vectors and study its properties in different gene-action scenarios. Two different kriging methods ["universal kriging" (UK) and "simple kriging" (SK)] are presented. As a novelty, we suggest use of the family of Matérn covariance functions to model the covariance structure of SNP vectors. A genomic best linear unbiased prediction (GBLUP) is applied as a reference method. The three approaches are compared in a whole-genome simulation study considering additive, additive-dominance, and epistatic gene-action models. Predictive performance is measured in terms of correlation between true and predicted GVs and average true GVs of the individuals ranked best by prediction. We show that UK outperforms GBLUP in the presence of dominance and epistatic effects. In a limiting case, it is shown that the genomic covariance structure proposed by VanRaden (2008) can be considered as a covariance function with corresponding quadratic variogram. We also prove theoretically that if a specific linear relationship exists between covariance matrices for two linear mixed models, the GVs resulting from BLUP are linked by a scaling factor. Finally, the relation of kriging to other models is discussed and further options for modeling the covariance structure, which might be more appropriate in the genomic context, are suggested.

Long-term impacts of genome-enabled selection.

The objective was to evaluate the effects of directional selection based on estimated genomic breeding values (GEBVs) for a quantitative trait. Selection affects GEBV prediction accuracy as well as genetic architecture via changes in allelic frequencies and linkage disequilibrium (LD), and the resulting changes are different from those in the absence of selection. How marker density affects long-term GEBV accuracy and selection response needs to be understood as well. Simulations were used to characterize the impact of selection based on GEBVs over generations. Single-nucleotide polymorphism (SNP) marker effects were estimated with the Bayesian Lasso method in the base generation, and these estimates were used to calculate the GEBVs in subsequent generations. GEBV accuracy decreased over generations of selection, and it was lower than under random selection, where a decay took place as well. In the long term, selection response tended to reach a plateau, but, at higher marker density, both the magnitude and duration of the response were larger. Selection changed quantitative trait loci (QTL) allele frequencies and generated new but unfavorable LD for prediction. Family effects had a considerable contribution to GEBV accuracy in early generations of selection.

Reconstruction of ancient genome and gene order from complete microbial genome sequences.

Microbial genome sequences provide us with the fossil records for inferring their origination and evolution. Assuming that current microbial genomes are the evolutionary results of ancient genomes or fragments and the neighboring genes in ancient genomes are more likely neighbors in current genomes, in this paper we proposed a paleontological algorithm and assembled the orthologous gene groups from 66 complete and current microbial genome sequences into a pseudo-ancient genome, which consists of continuous fragments of various sizes. We performed bootstrap resampling and correlation analyses and the results showed that the assembled ancient genome and fragments are statistically significant and the genes of the same fragment are inherently related and likely derived from common ancestors. This method provides a new computational tool for studying microbial genome structure and evolution.