Group-based variant calling leveraging next-generation supercomputing for large-scale whole-genome sequencing studies

Motivation

Next-generation sequencing (NGS) technologies have become much more efficient, allowing whole human genomes to be sequenced faster and cheaper than ever before. However, processing the raw sequence reads associated with NGS technologies requires care and sophistication in order to draw compelling inferences about phenotypic consequences of variation in human genomes. It has been shown that different approaches to variant calling from NGS data can lead to different conclusions. Ensuring appropriate accuracy and quality in variant calling can come at a computational cost.

Results

We describe our experience implementing and evaluating a group-based approach to calling variants on large numbers of whole human genomes. We explore the influence of many factors that may impact the accuracy and efficiency of group-based variant calling, including group size, the biogeographical backgrounds of the individuals who have been sequenced, and the computing environment used. We make efficient use of the Gordon supercomputer cluster at the San Diego Supercomputer Center by incorporating job-packing and parallelization considerations into our workflow while calling variants on 437 whole human genomes generated as part of large association study.

Conclusions

We ultimately find that our workflow resulted in high-quality variant calls in a computationally efficient manner. We argue that studies like ours should motivate further investigations combining hardware-oriented advances in computing systems with algorithmic developments to tackle emerging ‘big data’ problems in biomedical research brought on by the expansion of NGS technologies.

Electronic supplementary material

The online version of this article (doi:10.1186/s12859-015-0736-4) contains supplementary material, which is available to authorized users.     

Weight calibration to improve the efficiency of pure risk estimates from case-control samples nested in a cohort

Cohort studies provide information on relative hazards and pure risks of disease. For rare outcomes, large cohorts are needed to have sufficient numbers of events, making it costly to obtain covariate information on all cohort members. We focus on nested case-control designs that are used to estimate relative hazard in the Cox regression model. In 1997, Langholz and Borgan showed that pure risk can also be estimated from nested case-control data. However, these approaches do not take advantage of some covariates that may be available on all cohort members. Researchers have used weight calibration to increase the efficiency of relative hazard estimates from case-cohort studies and nested cased-control studies. Our objective is to extend weight calibration approaches to nested case-control designs to improve precision of estimates of relative hazards and pure risks. We show that calibrating sample weights additionally against follow-up times multiplied by relative hazards during the risk projection period improves estimates of pure risk. Efficiency improvements for relative hazards for variables that are available on the entire cohort also contribute to improved efficiency for pure risks. We develop explicit variance formulas for the weight-calibrated estimates. Simulations show how much precision is improved by calibration and confirm the validity of inference based on asymptotic normality. Examples are provided using data from the American Association of Retired Persons Diet and Health Cohort Study.     

Structural Insights into the Affinity of Cel7A Carbohydrate-binding Module for Lignin

The high cost of hydrolytic enzymes impedes the commercial production of lignocellulosic biofuels. High enzyme loadings are required in part due to their non-productive adsorption to lignin, a major component of biomass. Despite numerous studies documenting cellulase adsorption to lignin, few attempts have been made to engineer enzymes to reduce lignin binding. In this work, we used alanine-scanning mutagenesis to elucidate the structural basis for the lignin affinity of Trichoderma reesei Cel7A carbohydrate binding module (CBM). T. reesei Cel7A CBM mutants were produced with a Talaromyces emersonii Cel7A catalytic domain and screened for their binding to cellulose and lignin. Mutation of aromatic and polar residues on the planar face of the CBM greatly decreased binding to both cellulose and lignin, supporting the hypothesis that the cellulose-binding face is also responsible for lignin affinity. Cellulose and lignin affinity of the 31 mutants were highly correlated, although several mutants displayed selective reductions in lignin or cellulose affinity. Four mutants with increased cellulose selectivity (Q2A, H4A, V18A, and P30A) did not exhibit improved hydrolysis of cellulose in the presence of lignin. Further reduction in lignin affinity while maintaining a high level of cellulose affinity is thus necessary to generate an enzyme with improved hydrolysis capability. This work provides insights into the structural underpinnings of lignin affinity, identifies residues amenable to mutation without compromising cellulose affinity, and informs engineering strategies for family one CBMs.     

Evolution of linear chromosomes and multipartite genomes in yeast mitochondria

Mitochondrial genome diversity in closely related species provides an excellent platform for investigation of chromosome architecture and its evolution by means of comparative genomics. In this study, we determined the complete mitochondrial DNA sequences of eight Candida species and analyzed their molecular architectures. Our survey revealed a puzzling variability of genome architecture, including circular- and linear-mapping and multipartite linear forms. We propose that the arrangement of large inverted repeats identified in these genomes plays a crucial role in alterations of their molecular architectures. In specific arrangements, the inverted repeats appear to function as resolution elements, allowing genome conversion among different topologies, eventually leading to genome fragmentation into multiple linear DNA molecules. We suggest that molecular transactions generating linear mitochondrial DNA molecules with defined telomeric structures may parallel the evolutionary emergence of linear chromosomes and multipartite genomes in general and may provide clues for the origin of telomeres and pathways implicated in their maintenance.     

Paramount levels of ergothioneine transporter SLC22A4 mRNA in boar seminal vesicles and cross-species analysis of ergothioneine and glutathione in seminal plasma

Ergothioneine (ET) is a unique natural antioxidant which mammalia acquire exclusively from their food. Recently, we have discovered an ET transporter (ETT; gene symbol SLC22A4). The existence of a specific transporter suggests a beneficial role for ET; however, the precise physiological purpose of ET is still unclear. A conspicuous site of high extracellular ET accumulation is boar seminal plasma. Here, we have investigated whether ETT is responsible for specific accumulation of ET in the boar reproductive tract. The putative ETT from pig (ETTp) was cloned and validated by functional expression in 293 cells. The highest levels of ETTp mRNA were detected by real-time RT-PCR in seminal vesicles, eye, and kidney; much less was present in bulbourethral gland, testis, and prostate. By contrast, there was virtually no ETT mRNA in rat seminal vesicles. ET content in boar reproductive tissues, determined by LC-MS/MS, closely matched the ETT expression profile. Thus, strong and specific expression of ETTp in boar seminal vesicles explains high accumulation of ET in this gland and hence also in seminal plasma. Previous reports suggest that the glutathione (GSH) content of seminal plasma correlates directly with ET content; however, a comprehensive analysis across several species is not available. We have measured ET and GSH in seminal plasma from human, boar, bull, stallion, and rabbit by LC-MS/MS. GSH levels in seminal plasma do not correlate with ET levels. This suggests that the function of ET, at least in this extracellular context, does not depend on redox cycling with GSH.     

Mechanisms controlling virulence thresholds of mixed viral populations

The propensity of RNA viruses to revert attenuating mutations contributes to disease and complicates vaccine development. Despite the presence of virulent revertant viruses in some live-attenuated vaccines, disease from vaccination is rare. This suggests that in mixed viral populations, attenuated viruses may limit the pathogenesis of virulent viruses, thus establishing a virulence threshold. Here we examined virulence thresholds using mixtures of virulent and attenuated viruses in a transgenic mouse model of poliovirus infection. We determined that a 1,000-fold excess of the attenuated Sabin strain of poliovirus was protective against disease induced by the virulent Mahoney strain. Protection was induced locally, and inactivated virus conferred protection. Treatment with a poliovirus receptor-blocking antibody phenocopied the protective effect of inactivated viruses in vitro and in vivo, suggesting that one mechanism controlling virulence thresholds may be competition for a viral receptor. Additionally, the type I interferon response reduces poliovirus pathogenesis; therefore, we examined virulence thresholds in mice lacking the alpha/beta interferon receptor. We found that the attenuated virus was virulent in immunodeficient mice due to the enhanced replication and reversion of attenuating mutations. Therefore, while the type I interferon response limits the virulence of the attenuated strain by reducing replication, protection from disease conferred by the attenuated strain in immunocompetent mice can occur independently of replication. Our results identified mechanisms controlling the virulence of mixed viral populations and indicate that live-attenuated vaccines containing virulent virus may be safe, as long as virulent viruses are present at levels below a critical threshold.     

Synthesis and Characterization of Acceptor‐Substituted Oligothiophenes for Solar Cell Applications

We report the synthesis of novel acceptor‐substituted oligothiophenes and their application in m‐i‐p type planar heterojunction solar cells. Optical absorption spectra and electrochemical properties of the dyes are investigated. The determined energy levels of these dyes suggest that they should be ideal for use in heterojunction solar cells. We further investigate the influence of acceptor groups on the device performance by introducing 1,1‐dicyano‐2‐methyl‐vinyl and 1,1‐dicyano‐2‐phenyl‐vinyl groups, respectively, as acceptor units. Photovoltaic devices incorporating these dyes show an open circuit voltage of up to 0.96 V and power conversion efficiencies in the range of 1.5–3.0% under full sun illumination (simulated AM 1.5G sunlight, 100 mW cm^?2).     

BPEX Pig Health Scheme: a useful monitoring system for respiratory disease control in pig farms?

Background

Both O157 and non-O157 Shiga toxin - producing Escherichia coli (STECs) cause serious human disease outbreaks through the consumption of contaminated foods. Cattle are considered the main reservoir but it is unclear how STECs affect mature animals. Neonatal calves are the susceptible age class for STEC infections causing severe enteritis. In an earlier study, we determined that mycotoxins and STECs were part of the disease complex for dairy cattle with Jejunal Hemorrhage Syndrome (JHS). For STECs to play a role in the development of JHS, we hypothesized that STEC colonization should also be evident in beef cattle with JHS. Aggressive medical and surgical therapies are effective for JHS, but rely on early recognition of clinical signs for optimal outcomes suggesting that novel approaches must be developed for managing this disease. The main objective of this study was to confirm that mouldy feeds, mycotoxins and STEC colonization were associated with the development of JHS in beef cattle.

Results

Beef cattle developed JHS after consuming feed containing several types of mycotoxigenic fungi including Fusarium poae, F. verticillioides, F. sporotrichioides, Penicillium roqueforti and Aspergillus fumigatus. Mixtures of STECs colonized the mucosa in the hemorrhaged tissues of the cattle and no other pathogen was identified. The STECs expressed Stx1 and Stx2, but more significantly, Stxs were also present in the blood collected from the lumen of the hemorrhaged jejunum. Feed extracts containing mycotoxins were toxic to enterocytes and 0.1% of a prebiotic, Celmanax Trademark, removed the cytotoxicity in vitro. The inclusion of a prebiotic in the care program for symptomatic beef calves was associated with 69% recovery.

Conclusions

The current study confirmed that STECs and mycotoxins are part of the disease complex for JHS in beef cattle. Mycotoxigenic fungi are only relevant in that they produce the mycotoxins deposited in the feed. A prebiotic, Celmanax Trademark, acted as a mycotoxin binder in vitro and interfered with the progression of disease.     

An IL28B genotype-based clinical prediction model for treatment of chronic hepatitis C

Background

Genetic variation in IL28B and other factors are associated with sustained virological response (SVR) after pegylated-interferon/ribavirin treatment for chronic hepatitis C (CHC). Using data from the HALT-C Trial, we developed a model to predict a patient''s probability of SVR based on IL28B genotype and clinical variables.

Methods

HALT-C enrolled patients with advanced CHC who had failed previous interferon-based treatment. Subjects were re-treated with pegylated-interferon/ribavirin during trial lead-in. We used step-wise logistic regression to calculate adjusted odds ratios (aOR) and create the predictive model. Leave-one-out cross-validation was used to predict a priori probabilities of SVR and determine area under the receiver operator characteristics curve (AUC).

Results

Among 646 HCV genotype 1-infected European American patients, 14.2% achieved SVR. IL28B rs12979860-CC genotype was the strongest predictor of SVR (aOR, 7.56; p<.0001); the model also included HCV RNA (log10 IU/ml), AST∶ALT ratio, Ishak fibrosis score and prior ribavirin treatment. For this model AUC was 78.5%, compared to 73.0% for a model restricted to the four clinical predictors and 60.0% for a model restricted to IL28B genotype (p<0.001). Subjects with a predicted probability of SVR <10% had an observed SVR rate of 3.8%; subjects with a predicted probability >10% (43.3% of subjects) had an SVR rate of 27.9% and accounted for 84.8% of subjects actually achieving SVR. To verify that consideration of both IL28B genotype and clinical variables is required for treatment decisions, we calculated AUC values from published data for the IDEAL Study.

Conclusion

A clinical prediction model based on IL28B genotype and clinical variables can yield useful individualized predictions of the probability of treatment success that could increase SVR rates and decrease the frequency of futile treatment among patients with CHC.