Critical assessment of variant prioritization methods for rare disease diagnosis within the rare genomes project.

BACKGROUND: A major obstacle faced by families with rare diseases is obtaining a genetic diagnosis. The average "diagnostic odyssey" lasts over five years and causal variants are identified in under 50%, even when capturing variants genome-wide. To aid in the interpretation and prioritization of the vast number of variants detected, computational methods are proliferating. Knowing which tools are most effective remains unclear. To evaluate the performance of computational methods, and to encourage innovation in method development, we designed a Critical Assessment of Genome Interpretation (CAGI) community challenge to place variant prioritization models head-to-head in a real-life clinical diagnostic setting. METHODS: We utilized genome sequencing (GS) data from families sequenced in the Rare Genomes Project (RGP), a direct-to-participant research study on the utility of GS for rare disease diagnosis and gene discovery. Challenge predictors were provided with a dataset of variant calls and phenotype terms from 175 RGP individuals (65 families), including 35 solved training set families with causal variants specified, and 30 unlabeled test set families (14 solved, 16 unsolved). We tasked teams to identify causal variants in as many families as possible. Predictors submitted variant predictions with estimated probability of causal relationship (EPCR) values. Model performance was determined by two metrics, a weighted score based on the rank position of causal variants, and the maximum F-measure, based on precision and recall of causal variants across all EPCR values. RESULTS: Sixteen teams submitted predictions from 52 models, some with manual review incorporated. Top performers recalled causal variants in up to 13 of 14 solved families within the top 5 ranked variants. Newly discovered diagnostic variants were returned to two previously unsolved families following confirmatory RNA sequencing, and two novel disease gene candidates were entered into Matchmaker Exchange. In one example, RNA sequencing demonstrated aberrant splicing due to a deep intronic indel in ASNS, identified in trans with a frameshift variant in an unsolved proband with phenotypes consistent with asparagine synthetase deficiency. CONCLUSIONS: Model methodology and performance was highly variable. Models weighing call quality, allele frequency, predicted deleteriousness, segregation, and phenotype were effective in identifying causal variants, and models open to phenotype expansion and non-coding variants were able to capture more difficult diagnoses and discover new diagnoses. Overall, computational models can significantly aid variant prioritization. For use in diagnostics, detailed review and conservative assessment of prioritized variants against established criteria is needed.

Critical assessment of variant prioritization methods for rare disease diagnosis within the Rare Genomes Project.

Background: A major obstacle faced by rare disease families is obtaining a genetic diagnosis. The average "diagnostic odyssey" lasts over five years, and causal variants are identified in under 50%. The Rare Genomes Project (RGP) is a direct-to-participant research study on the utility of genome sequencing (GS) for diagnosis and gene discovery. Families are consented for sharing of sequence and phenotype data with researchers, allowing development of a Critical Assessment of Genome Interpretation (CAGI) community challenge, placing variant prioritization models head-to-head in a real-life clinical diagnostic setting. Methods: Predictors were provided a dataset of phenotype terms and variant calls from GS of 175 RGP individuals (65 families), including 35 solved training set families, with causal variants specified, and 30 test set families (14 solved, 16 unsolved). The challenge tasked teams with identifying the causal variants in as many test set families as possible. Ranked variant predictions were submitted with estimated probability of causal relationship (EPCR) values. Model performance was determined by two metrics, a weighted score based on rank position of true positive causal variants and maximum F-measure, based on precision and recall of causal variants across EPCR thresholds. Results: Sixteen teams submitted predictions from 52 models, some with manual review incorporated. Top performing teams recalled the causal variants in up to 13 of 14 solved families by prioritizing high quality variant calls that were rare, predicted deleterious, segregating correctly, and consistent with reported phenotype. In unsolved families, newly discovered diagnostic variants were returned to two families following confirmatory RNA sequencing, and two prioritized novel disease gene candidates were entered into Matchmaker Exchange. In one example, RNA sequencing demonstrated aberrant splicing due to a deep intronic indel in ASNS, identified in trans with a frameshift variant, in an unsolved proband with phenotype overlap with asparagine synthetase deficiency. Conclusions: By objective assessment of variant predictions, we provide insights into current state-of-the-art algorithms and platforms for genome sequencing analysis for rare disease diagnosis and explore areas for future optimization. Identification of diagnostic variants in unsolved families promotes synergy between researchers with clinical and computational expertise as a means of advancing the field of clinical genome interpretation.

Partially automated whole-genome sequencing reanalysis of previously undiagnosed pediatric patients can efficiently yield new diagnoses.

To investigate the diagnostic and clinical utility of a partially automated reanalysis pipeline, forty-eight cases of seriously ill children with suspected genetic disease who did not receive a diagnosis upon initial manual analysis of whole-genome sequencing (WGS) were reanalyzed at least 1 year later. Clinical natural language processing (CNLP) of medical records provided automated, updated patient phenotypes, and an automated analysis system delivered limited lists of possible diagnostic variants for each case. CNLP identified a median of 79 new clinical features per patient at least 1 year later. Compared to a standard manual reanalysis pipeline, the partially automated pipeline reduced the number of variants to be analyzed by 90% (range: 74%-96%). In 2 cases, diagnoses were made upon reinterpretation, representing an incremental diagnostic yield of 4.2% (2/48, 95% CI: 0.5-14.3%). Four additional cases were flagged with a possible diagnosis to be considered during subsequent reanalysis. Separately, copy number analysis led to diagnoses in two cases. Ongoing discovery of new disease genes and refined variant classification necessitate periodic reanalysis of negative WGS cases. The clinical features of patients sequenced as infants evolve rapidly with age. Partially automated reanalysis, including automated re-phenotyping through CNLP, has the potential to identify molecular diagnoses with reduced expert labor intensity.

A Randomized, Controlled Trial of the Analytic and Diagnostic Performance of Singleton and Trio, Rapid Genome and Exome Sequencing in Ill Infants.

The second Newborn Sequencing in Genomic Medicine and Public Health study was a randomized, controlled trial of the effectiveness of rapid whole-genome or -exome sequencing (rWGS or rWES, respectively) in seriously ill infants with diseases of unknown etiology. Here we report comparisons of analytic and diagnostic performance. Of 1,248 ill inpatient infants, 578 (46%) had diseases of unknown etiology. 213 infants (37% of those eligible) were enrolled within 96 h of admission. 24 infants (11%) were very ill and received ultra-rapid whole-genome sequencing (urWGS). The remaining infants were randomized, 95 to rWES and 94 to rWGS. The analytic performance of rWGS was superior to rWES, including variants likely to affect protein function, and ClinVar pathogenic/likely pathogenic variants (p < 0.0001). The diagnostic performance of rWGS and rWES were similar (18 diagnoses in 94 infants [19%] versus 19 diagnoses in 95 infants [20%], respectively), as was time to result (median 11.0 versus 11.2 days, respectively). However, the proportion diagnosed by urWGS (11 of 24 [46%]) was higher than rWES/rWGS (p = 0.004) and time to result was less (median 4.6 days, p < 0.0001). The incremental diagnostic yield of reflexing to trio after negative proband analysis was 0.7% (1 of 147). In conclusion, rapid genomic sequencing can be performed as a first-tier diagnostic test in inpatient infants. urWGS had the shortest time to result, which was important in unstable infants, and those in whom a genetic diagnosis was likely to impact immediate management. Further comparison of urWGS and rWES is warranted because genomic technologies and knowledge of variant pathogenicity are evolving rapidly.

Biochemical determinants of ObgE-mediated persistence.

Obg is a versatile GTPase that plays a pivotal role in bacterial persistence. We previously showed that the Escherichia coli homolog ObgE exerts this activity through transcriptional activation of a toxin-antitoxin module and subsequent membrane depolarization. Here, we assessed the role of G-domain functionality in ObgE-mediated persistence. Through screening of a mutant library, we identified five obgE alleles (with substitutions G166V, D246G, S270I, N283I and I313N) that have lost their persistence function and no longer activate hokB expression. These alleles support viability of a strain otherwise deprived of ObgE, indicating that ObgE's persistence function can be uncoupled from its essential role. Based on the ObgE crystal structure, we designed two additional mutant proteins (T193A and D286Y), one of which (D286Y) no longer affects persistence. Using isothermal titration calorimetry, stopped-flow experiments and kinetics, we subsequently assessed nucleotide binding and GTPase activity in all mutants. With the exception of the S270I mutant that is possibly affected in protein-protein interactions, all mutants that have lost their persistence function display severely reduced binding to GDP or the alarmone ppGpp. However, we find no clear relation between persistence and GTP or pppGpp binding nor with GTP hydrolysis. Combined, our results signify an important step toward understanding biochemical determinants underlying persistence.

Diagnosis of genetic diseases in seriously ill children by rapid whole-genome sequencing and automated phenotyping and interpretation.

By informing timely targeted treatments, rapid whole-genome sequencing can improve the outcomes of seriously ill children with genetic diseases, particularly infants in neonatal and pediatric intensive care units (ICUs). The need for highly qualified professionals to decipher results, however, precludes widespread implementation. We describe a platform for population-scale, provisional diagnosis of genetic diseases with automated phenotyping and interpretation. Genome sequencing was expedited by bead-based genome library preparation directly from blood samples and sequencing of paired 100-nt reads in 15.5 hours. Clinical natural language processing (CNLP) automatically extracted children's deep phenomes from electronic health records with 80% precision and 93% recall. In 101 children with 105 genetic diseases, a mean of 4.3 CNLP-extracted phenotypic features matched the expected phenotypic features of those diseases, compared with a match of 0.9 phenotypic features used in manual interpretation. We automated provisional diagnosis by combining the ranking of the similarity of a patient's CNLP phenome with respect to the expected phenotypic features of all genetic diseases, together with the ranking of the pathogenicity of all of the patient's genomic variants. Automated, retrospective diagnoses concurred well with expert manual interpretation (97% recall and 99% precision in 95 children with 97 genetic diseases). Prospectively, our platform correctly diagnosed three of seven seriously ill ICU infants (100% precision and recall) with a mean time saving of 22:19 hours. In each case, the diagnosis affected treatment. Genome sequencing with automated phenotyping and interpretation in a median of 20:10 hours may increase adoption in ICUs and, thereby, timely implementation of precise treatments.

A Single-Amino-Acid Substitution in Obg Activates a New Programmed Cell Death Pathway in Escherichia coli.

UNLABELLED: Programmed cell death (PCD) is an important hallmark of multicellular organisms. Cells self-destruct through a regulated series of events for the benefit of the organism as a whole. The existence of PCD in bacteria has long been controversial due to the widely held belief that only multicellular organisms would profit from this kind of altruistic behavior at the cellular level. However, over the past decade, compelling experimental evidence has established the existence of such pathways in bacteria. Here, we report that expression of a mutant isoform of the essential GTPase ObgE causes rapid loss of viability in Escherichia coli. The physiological changes that occur upon expression of this mutant protein--including loss of membrane potential, chromosome condensation and fragmentation, exposure of phosphatidylserine on the cell surface, and membrane blebbing--point to a PCD mechanism. Importantly, key regulators and executioners of known bacterial PCD pathways were shown not to influence this cell death program. Collectively, our results suggest that the cell death pathway described in this work constitutes a new mode of bacterial PCD. IMPORTANCE: Programmed cell death (PCD) is a well-known phenomenon in higher eukaryotes. In these organisms, PCD is essential for embryonic development--for example, the disappearance of the interdigital web--and also functions in tissue homeostasis and elimination of pathogen-invaded cells. The existence of PCD mechanisms in unicellular organisms like bacteria, on the other hand, has only recently begun to be recognized. We here demonstrate the existence of a bacterial PCD pathway that induces characteristics that are strikingly reminiscent of eukaryotic apoptosis, such as fragmentation of DNA, exposure of phosphatidylserine on the cell surface, and membrane blebbing. Our results can provide more insight into the mechanism and evolution of PCD pathways in higher eukaryotes. More importantly, especially in the light of the looming antibiotic crisis, they may point to a bacterial Achilles' heel and can inspire innovative ways of combating bacterial infections, directed at the targeted activation of PCD pathways.

Obg and Membrane Depolarization Are Part of a Microbial Bet-Hedging Strategy that Leads to Antibiotic Tolerance.

Within bacterial populations, a small fraction of persister cells is transiently capable of surviving exposure to lethal doses of antibiotics. As a bet-hedging strategy, persistence levels are determined both by stochastic induction and by environmental stimuli called responsive diversification. Little is known about the mechanisms that link the low frequency of persisters to environmental signals. Our results support a central role for the conserved GTPase Obg in determining persistence in Escherichia coli in response to nutrient starvation. Obg-mediated persistence requires the stringent response alarmone (p)ppGpp and proceeds through transcriptional control of the hokB-sokB type I toxin-antitoxin module. In individual cells, increased Obg levels induce HokB expression, which in turn results in a collapse of the membrane potential, leading to dormancy. Obg also controls persistence in Pseudomonas aeruginosa and thus constitutes a conserved regulator of antibiotic tolerance. Combined, our findings signify an important step toward unraveling shared genetic mechanisms underlying persistence.

A putative de-N-acetylase of the PIG-L superfamily affects fluoroquinolone tolerance in Pseudomonas aeruginosa.

A major cause of treatment failure of infections caused by Pseudomonas aeruginosa is the presence of antibiotic-insensitive persister cells. The mechanism of persister formation in P. aeruginosa is largely unknown, and so far, only few genetic determinants have been linked to P. aeruginosa persistence. Based on a previous high-throughput screening, we here present dnpA (de-N-acetylase involved in persistence; gene locus PA14_66140/PA5002) as a new gene involved in noninherited fluoroquinolone tolerance in P. aeruginosa. Fluoroquinolone tolerance of a dnpA mutant is strongly reduced both in planktonic culture and in a biofilm model, whereas overexpression of dnpA in the wild-type strain increases the persister fraction. In addition, the susceptibility of the dnpA mutant to different classes of antibiotics is not affected. dnpA is part of the conserved LPS core oligosaccharide biosynthesis gene cluster. Based on primary sequence analysis, we predict that DnpA is a de-N-acetylase, acting on an unidentified substrate. Site-directed mutagenesis suggests that this enzymatic activity is essential for DnpA-mediated persistence. A transcriptome analysis indicates that DnpA primarily affects the expression of genes involved in surface-associated processes. We discuss the implications of these findings for future antipersister therapies targeted at chronic P. aeruginosa infections.

Bacterial Obg proteins: GTPases at the nexus of protein and DNA synthesis.

Obg proteins (also known as ObgE, YhbZ and CgtA) are conserved P-loop GTPases, essential for growth in bacteria. Like other GTPases, Obg proteins cycle between a GTP-bound ON and a GDP-bound OFF state, thereby controlling cellular processes. Interestingly, the in vitro biochemical properties of Obg proteins suggest that they act as sensors for the cellular GDP/GTP pools and adjust their activity according to the cellular energy status. Obg proteins have been attributed a host of cellular functions, including roles in essential cellular processes (DNA replication, ribosome maturation) and roles in different stress adaptation pathways (stringent response, sporulation, general stress response). This review summarizes the current knowledge on Obg activity and function. Furthermore, we present a model that integrates the different functions of Obg by assigning it a fundamental role in cellular physiology, at the hub of protein and DNA synthesis. In particular, we believe that Obg proteins might provide a connection between different global pathways in order to fine-tune cellular processes in response to a given energy status.

New-found fundamentals of bacterial persistence.

Persister cells display tolerance to high doses of bactericidal antibiotics and typically comprise a small fraction of a bacterial population. Recently, evidence was provided for a causal link between therapy failure and the presence of persister cells in chronic infections, underscoring the need for research on bacterial persistence. A series of recent breakthroughs have shed light on the multiplicity of persister genes, the contribution of gene expression noise to persister formation, the importance of active responses to antibiotic tolerance and heterogeneity among persister cells. Moreover, the development of in vivo model systems has highlighted the clinical relevance of persistence. This review discusses these recent advances and how this knowledge fundamentally changes the way in which we will perceive the problem of antibiotic tolerance in years to come.

The Escherichia coli GTPase ObgE modulates hydroxyl radical levels in response to DNA replication fork arrest.

Obg proteins are universally conserved GTP-binding proteins that are essential for viability in bacteria. Homologs in different organisms are involved in various cellular processes, including DNA replication. The goal of this study was to analyse the structure-function relationship of Escherichia coli ObgE with regard to DNA replication in general and sensitivity to stalled replication forks in particular. Defined C-terminal chromosomal deletion mutants of obgE were constructed and tested for sensitivity to the replication inhibitor hydroxyurea. The ObgE C-terminal domain was shown to be dispensable for normal growth of E.coli. However, a region within this domain is involved in the cellular response to replication fork stress. In addition, a mutant obgE over-expression library was constructed by error-prone PCR and screened for increased hydroxyurea sensitivity. ObgE proteins with substitutions L159Q, G163V, P168V, G216A or R237C, located within distinct domains of ObgE, display dominant-negative effects leading to hydroxyurea hypersensitivity when over-expressed. These effects are abolished in strains with a single deletion of the iron transporter TonB or combined deletions the toxin/antitoxin modules RelBE/MazEF, strains both of which have been shown to be involved in a pathway that stimulates hydroxyl radical formation following hydroxyurea treatment. Moreover, the observed dominant-negative effects are lost in the presence of the hydroxyl radical scavenger thiourea. Together, these results indicate involvement of hydroxyl radical toxicity in ObgE-mediated protection against replication fork stress.

Spectroscopic characterization of Venus at the single molecule level.

Venus is a recently developed, fast maturating, yellow fluorescent protein that has been used as a probe for in vivo applications. In the present work the photophysical characteristics of Venus were analyzed spectroscopically at the bulk and single molecule level. Through time-resolved single molecule measurements we found that single molecules of Venus display pronounced fluctuations in fluorescence emission, with clear fluorescence on- and off-times. These fluorescence intermittencies were found to occupy a broad range of time scales, ranging from milliseconds to several seconds. Such long off-times can complicate the analysis of single molecule counting experiments or single-molecule FRET experiments.

Pseudomonas aeruginosa fosfomycin resistance mechanisms affect non-inherited fluoroquinolone tolerance.

Pseudomonas aeruginosa is an opportunistic pathogen that poses a threat in clinical settings due to its intrinsic and acquired resistance to a wide spectrum of antibiotics. Additionally, the presence of a subpopulation of cells surviving high concentrations of antibiotics, called persisters, makes it virtually impossible to eradicate a chronic infection. The mechanism underlying persistence is still unclear, partly due to the fact that it is a non-inherited phenotype. Based on our findings from a previously performed screening effort for P. aeruginosa persistence genes, we hypothesize that crosstalk can occur between two clinically relevant mechanisms: the persistence phenomenon and antibiotic resistance. This was tested by determining the persistence phenotype of P. aeruginosa strains that are resistant to the antibiotic fosfomycin due to either of two unrelated fosfomycin resistance mechanisms. Overexpression of fosA (PA1129) confers fosfomycin resistance by enzymic modification of the antibiotic, and in addition causes a decrease in the number of persister cells surviving ofloxacin treatment. Both phenotypes require the enzymic function of FosA, as mutation of the Arg119 residue abolishes fosfomycin resistance as well as low persistence. The role for fosfomycin resistance mechanisms in persistence is corroborated by demonstrating a similar phenotype in a strain with a mutation in glpT (PA5235), which encodes a glycerol-3-phosphate transporter essential for fosfomycin uptake. These results indicate that fosfomycin resistance, conferred by glpT mutation or by overexpression of fosA, results in a decrease in the number of persister cells after treatment with ofloxacin and additionally stress that further research into the interplay between fosfomycin resistance and persistence is warranted.

Novel persistence genes in Pseudomonas aeruginosa identified by high-throughput screening.

Persister cells are phenotypic variants that are extremely tolerant to high concentrations of antibiotics. They constitute a fraction of stationary phase cultures and biofilm populations of numerous bacterial species, such as the opportunistic pathogen Pseudomonas aeruginosa. Even though persisters are believed to be an important cause of incomplete elimination of infectious populations by antibiotics, their nature remains obscure. Most studies on persistence have focused on the model organism Escherichia coli and only a limited number of persistence genes have been identified to date. We performed the first large-scale screening of a P. aeruginosa PA14 mutant library to identify novel genes involved in persistence. A total of 5000 mutants were screened in a high-throughput manner and nine new persistence mutants were identified. Four mutants (with insertions in dinG, spuC, PA14_17880 and PA14_66140) exhibited a low persister phenotype and five mutants (in algR, pilH, ycgM, pheA and PA14_13680) displayed high persistence. These genes may serve as new candidate drug targets in the combat against P. aeruginosa infections.