Biology subjects

Cohen, T.

Publications and source records attributed to Cohen, T..

31 records · Page 2Linked to original sources

bioRxiv2020

vCOMBAT: a Novel Tool to Create and Visualize a COmputational Model of Bacterial Antibiotic Target-binding

MotivationAs antibiotic resistance creates a significant global health threat, we need not only to accelerate the development of novel antibiotics but also to develop better treatment strategies using existing drugs to improve their efficacy and prevent the selection of further resistance. We require new tools to rationally design dosing regimens to from data collected in early phases of antibiotic and dosing development. Mathematical models such as mechanistic pharmacodynamic drug-target binding explain mechanistic details of how the given drug concentration affects its targeted bacteria. However, there are no available tools in the literature that allows non-quantitative scientists to develop computational models to simulate antibiotic-target binding and its effects on bacteria. ResultsIn this work, we have devised an extension of a mechanistic binding-kinetic model to incorporate clinical drug concentration data. Based on the extended model, we develop a novel and interactive web-based tool that allows non-quantitative scientists to create and visualize their own computational models of bacterial antibiotic target-binding based on their considered drugs and bacteria. We also demonstrate how Rifampicin affects bacterial populations of Tuberculosis (TB) bacteria using our vCOMBAT tool. AvailabilityvCOMBAT online tool is publicly available at https://combat-bacteria.org/.

bioinformatics↗

bioRxiv2020

Mitochondrial and insulin gene expression in single cells shape pancreatic beta cells' population divergence

Mitochondrial gene expression is pivotal to cell metabolism. Nevertheless, it is unknown whether it diverges within a given cell type. Here, we analysed single-cell RNA-seq experiments from [~]4600 human pancreatic alpha and beta cells, as well as [~]900 mouse beta cells. Cluster analysis revealed two distinct human beta cells populations, which diverged by mitochondrial (mtDNA) and nuclear DNA (nDNA)-encoded oxidative phosphorylation (OXPHOS) gene expression in healthy and diabetic individuals, and in newborn but not in adult mice. Insulin gene expression was elevated in beta cells with higher mtDNA gene expression in humans and in young mice. Such human beta cell populations also diverged in mt-RNA mutational repertoire, and in their selective signature, thus implying the existence of two previously overlooked distinct and conserved beta cell populations. While applying our approach to alpha cells, two sub-populations of cells were identified which diverged in mtDNA gene expression, yet these cellular populations did not consistently diverge in nDNA OXPHOS genes expression, nor did they correlate with the expression of glucagon, the hallmark of alpha cells. Thus, pancreatic beta cells within an individual are divided into distinct groups with unique metabolic-mitochondrial signature.

genomics↗

bioRxiv2020

Mechanisms of antibiotic action shape the fitness landscapes of resistance mutations

Antibiotic-resistant pathogens are a major public health threat. A deeper understanding of how an antibiotics mechanism of action influences the emergence of resistance would aid in the design of new drugs and help to preserve the effectiveness of existing ones. To this end, we developed a model that links bacterial population dynamics with antibiotic-target binding kinetics. Our approach allows us to derive mechanistic insights on drug activity from population-scale experimental data and to quantify the interplay between drug mechanism and resistance selection. We find that whether a drug acts as a bacteriostatic or bactericidal agent has little influence on resistance selection. We also show that heterogeneous drug-target binding within a population enables resistant bacteria to evolve fitness-improving secondary mutations even when drug doses remain above the resistant strains minimum inhibitory concentration. Our work suggests that antibiotic doses beyond this "secondary mutation selection window" could safeguard against the emergence of high-fitness resistant strains during treatment.

pharmacology and toxicology↗

bioRxiv2019

aer2vec: Distributed Representations of Adverse Event Reporting System Data as a Means to Identify Drug/Side-Effect Associations

Adverse event report (AER) data are a key source of signal for post marketing drug surveillance. The standard methodology to analyze AER data applies disproportionality metrics, which estimate the strength of drug/side-effect associations from discrete counts of their occurrence at report level. However, in other domains, improvements in predictive modeling accuracy have been obtained through representation learning, where discrete features are replaced by distributed representations learned from unlabeled data. This paper describes aer2vec, a novel representational approach for AER data in which concept embeddings emerge from neural networks trained to predict drug/side-effect co-occurrence. Trained models are evaluated for their utility in identifying drug/side-effect relationships, with improvements over disproportionality metrics in most cases. In addition, we evaluate the utility of an otherwise-untapped resource in the Food and Drug Administration (FDA) AER system - reporter designations of suspected causality - and find that incorporating this information enhances performance of all models evaluated.

bioinformatics↗

bioRxiv2019

Predicting Adverse Drug-Drug Interactions with Neural Embedding of Semantic Predications

The identification of drug-drug interactions (DDIs) is important for patient safety; yet, compared to other pharmacovigilance work, a limited amount of research has been conducted in this space. Recent work has successfully applied a method of deriving distributed vector representations from structured biomedical knowledge, known as Embedding of Semantic Predications (ESP), to the problem of predicting individual drug side effects. In the current paper we extend this work by applying ESP to the problem of predicting polypharmacy side-effects for particular drug combinations, building on a recent reconceptualization of this problem as a network of drug nodes connected by side effect edges. We evaluate ESP embeddings derived from the resulting graph on a side-effect prediction task against a previously reported graph convolutional neural network approach, using the same data and evaluation methods. We demonstrate that ESP models perform better, while being faster to train, more re-usable, and significantly simpler.

bioinformatics↗

bioRxiv2019

Genomic variant identification methods alter Mycobacterium tuberculosis transmission inference

Pathogen genomic data are increasingly used to characterize global and local transmission patterns of important human pathogens and to inform public health interventions. Yet there is no current consensus on how to measure genomic variation. We investigated the effects of variant identification approaches on transmission inferences for M. tuberculosis by comparing variants identified by five different groups in the same sequence data from a clonal outbreak. We then measured the performance of commonly used variant calling approaches in recovering variation in a simulated tuberculosis outbreak and tested the effect of applying increasingly stringent filters on transmission inferences and phylogenies. We found that variant calling approaches used by different groups do not recover consistent sets of variants, often leading to conflicting transmission inferences. Further, performance in recovering true outbreak variation varied widely across approaches. Finally, stringent filters rapidly eroded the accuracy of transmission inferences and quality of phylogenies reconstructed from outbreak variation. We conclude that measurements of genetic distance and phylogenetic structure are dependent on variant calling approach. Variant calling algorithms trained upon true sequence data outperform other approaches and enable inclusion of repetitive regions typically excluded from genomic epidemiology studies, maximizing the information gleaned from outbreak genomes.

genomics↗

bioRxiv2019

Mutant MFN2/fzo-1 selectively removes mitochondrial DNA heteroplasmy in Caenorhabditis elegans and in Charcot-Marie-Tooth patients

Deleterious and intact mitochondrial DNA (mtDNA) mutations frequently co-exist (heteroplasmy). Such mutations likely survive and are inherited due to complementation via the intra-cellular mitochondrial network. Hence, we hypothesized that compromised mitochondrial fusion would hamper such complementation, thereby affecting heteroplasmy inheritance. To test this hypothesis, we assessed heteroplasmy levels in three Caenorhabditis elegans strains carrying different heteroplasmic mtDNA deletions ({Delta}mtDNA) in the background of mutant mitofusin. Firstly, these animals displayed severe embryonic lethality and developmental delay. Strikingly, these phenotypes were relieved during subsequent generations in association with complete {Delta}mtDNA removal. Moreover, the rates of deletion loss negatively correlated with the size of mtDNA deletions, suggesting that mitochondrial fusion is essential and sensitive to the nature of the heteroplasmic mtDNA mutations. While introducing the {Delta}mtDNA into a fzo-1;pdr-1 (PARKIN ortholog) double mutant, we observed skew in the mendelian distribution of progeny, in contrast to normal distribution in the {Delta}mtDNA;fzo-1 mutant, and severely reduced brood size. Notably, the {Delta}mtDNA was lost across generations in association with improved phenotypes. This underlines the importance of cross-talk between mitochondrial fusion and mitophagy in modulating the inheritance of mtDNA heteroplasmy. Finally, while investigating heteroplasmy patterns in three Charcot-Marie-Tooth disease type 2A pedigrees, which carry a mutated mitofusin 2, we found a single potentially deleterious heteroplasmic mutation, whose levels were selectively reduced in the patient versus healthy maternal relatives. Taken together our findings show that when mitochondrial fusion is compromised, deleterious heteroplasmic mutations cannot evade natural selection, while inherited from generation to generation.

genetics↗

bioRxiv2018

mtDNA eQTLs and the m1A 16S rRNA modification explain mtDNA tissue-specific gene expression pattern in humans

Expression quantitative trait loci (eQTLs) are instrumental in genome-wide identification of regulatory elements, yet were overlooked in the mitochondrial DNA (mtDNA). By analyzing 5079 RNA-seq samples from 23 tissues we identified association of ancient mtDNA SNPs (haplogroups T2, L2, J2 and V) and recurrent SNPs (mtDNA positions 263, 750, 1438 and 10398) with tissue-dependent mtDNA gene-expression. Since the recurrent SNPs independently occurred in different mtDNA genetic backgrounds, they constitute the best candidates to be causal eQTLs. Secondly, the discovery of mtDNA eQTLs in both coding and non-coding mtDNA regions, propose the identification of novel mtDNA regulatory elements. Third, we identified association between low m1A 947 MT-RNR2 (16S) rRNA modification levels and altered mtDNA gene-expression in twelve tissues. Such association disappeared in skin which was exposed to sun, as compared to sun-unexposed skin from the same individuals, thus supporting the impact of UV on mtDNA gene expression. Taken together, our findings reveal that both mtDNA SNPs and mt-rRNA modification affect mtDNA gene expression in a tissue-dependent manner.

genomics↗

bioRxiv2018

RESISTANCE AT NO COST: THE TRANSMISSIBILITY AND POTENTIAL FOR DISEASE PROGRESSION OF DRUG-RESISTANT M. TUBERCULOSIS

BackgroundThe future trajectory of drug resistant tuberculosis strongly depends on the fitness costs of drug resistance mutations. Here, we measured the association of phenotypic drug resistance and the risk of TB infection and disease among household contacts (HHCs) of patients with pulmonary TB.\n\nMethodsWe evaluated 12767 HHCs of patients with drug sensitive and resistant pulmonary TB at baseline, two, six, and 12 months to ascertain infection status and to determine whether they developed tuberculosis disease. We also assessed the impact of drug resistance phenotype on the likelihood that a TB strain shared a genetic fingerprint with at least one other TB patient in the cohort.\n\nFindingsAmong 3339 TB patients for whom were DST available, 1274 (38%) had TB that was resistant to at least one drug and 478 (14*3%) had multi-drug resistant (MDR) TB, i.e. TB resistant to both INH and rifampicin. Compared to HHCs of drug sensitive TB patients, those exposed to a patient with MDR-TB had an 8% (95% CI: 4-13%) higher risk of infection by the end of follow up. We found no statistically significant difference in the relative hazard of incident TB disease among HHCs exposed to MDR-TB compared to DS-TB (Adjusted HR 1*28 [(95% CI: *9-1*83]). Patients with MDR-TB were more likely to be part of a genetic cluster than were DS-TB patients.\n\nInterpretationClinical strains of MDR M. tuberculosis are neither less transmissible than drug sensitive strains nor less likely to cause disease. (ClinicalTrials.gov number, NCT00676754)\n\nFundingNational Institutes of Health: NIH/NIAID CETR U19AI109755\n\nStatementAll authors have seen the manuscript and approved the manuscript.

epidemiology↗

bioRxiv2018

Evaluation of tuberculosis treatment response with serial C-reactive protein measurements

BackgroundNovel biomarkers are needed to assess response to antituberculosis therapy in smear-negative patients.\n\nMethodsTo evaluate the utility of CRP in monitoring response to antituberculosis therapy we conducted a post-hoc analysis on a cohort of adults with symptoms of tuberculosis and negative sputum smears in a high tuberculosis and HIV prevalence setting in KwaZulu-Natal, South Africa. Serial changes in CRP, weight, and haemoglobin were evaluated over 8 weeks.\n\nResults421 participants with suspected smear-negative tuberculosis were enrolled and 33 excluded. 295 were treated for tuberculosis (137 confirmed, 158 possible), and 93 did not have tuberculosis. 185 of 215 (86%) participants who agreed to HIV testing were HIV-positive. At week 8, the on-treatment median CRP reduction in the tuberculosis group was 79.5% (IQR 25.4; 91.7), median weight gain 2.3% (IQR -1.0; 5.6), and median haemoglobin increase 7.0% (IQR 0.8; 18.9) (p-value <0.0001 for baseline to week 8 comparison of absolute median values). Only CRP changed significantly at week 2 (median reduction of 75.1% (IQR 46.9; 89.2) in the group with confirmed tuberculosis and 49.0% (IQR -0.4; 80.9) in the possible tuberculosis group. Failure of CRP to reduce to [≤]55% of the baseline value at week 2 predicted hospitalization or death in both tuberculosis groups, with 99% negative predictive value.\n\nConclusionChange in CRP may have utility in early evaluation of response to antituberculosis treatment and to identify those at increased risk of adverse outcomes.\n\nKey pointsC-reactive protein (CRP) falls by 80% after eight weeks of antituberculosis treatment. At two weeks sustained CRP elevation is associated with death or hospitalization.

microbiology↗

bioRxiv2018

How molecular mechanisms of resistance affect resistance evolution

Combatting antibiotic resistance will require both new antibiotics and strategies to preserve the effectiveness of existing drugs. Both approaches would benefit from predicting optimal dosing of antibiotics based on drug-target binding parameters that can be measured early in drug development and that can change when bacteria become resistant. This would avoid the currently frequently employed trial-and-error approaches and might reduce the number of antibiotic candidates that fail late in drug development.\n\nHere, we describe a computational model (COMBAT-COmputational Model of Bacterial Antibiotic Target-binding) that leverages accessible biochemical parameters to quantitatively predict antibiotic dose-response relationships. We validate our model with MICs of a range of quinolone antibiotics in clinical isolates demonstrating that antibiotic efficacy can be predicted from drug-target binding (R2 > 0.9). To further challenge our approach, we do not only predict antibiotic efficacy from biochemical parameters, but also do the reverse: estimate the magnitude of changes in drug-target binding based on antibiotic dose-response curves. We experimentally demonstrate that changes in drug-target binding can be predicted from antibiotic dose-response curves with 92-94 % accuracy by exposing bacteria overexpressing target molecules to ciprofloxacin. To test the generality of COMBAT, we apply it to a different antibiotic class, the beta-lactam ampicillin, and can again predict binding parameters from dose-response curves with 90 % accuracy. We then apply COMBAT to predict antibiotic concentrations that can select for resistance due to novel resistance mutations.\n\nOur goal here is dual: First, we address a fundamental biological question and demonstrate that drug-target binding determines bacterial response to antibiotics, although antibiotic action involves many additional effects downstream of drug-target binding. Second, we create a tool that can help accelerate drug development by predicting optimal dosing and preserve the efficacy of existing antibiotics by predicting optimal treatment for possible resistant mutants.

biochemistry↗

bioRxiv2018

Beyond the SNP threshold: identifying outbreak clusters using inferred transmissions

Whole genome sequencing (WGS) is increasingly used to aid in understanding pathogen transmission [1]. Very often the number of single nucleotide polymorphisms (SNPs) separating isolates collected during an epidemiological study are used to identify sets of cases that are potentially linked by direct transmission. However, there is little agreement in the literature as to what an appropriate SNP cut-off threshold should be, or indeed whether a simple SNP threshold is appropriate for identifying sets of isolates to be treated as \"transmission clusters\". The SNP thresholds that have been adopted for inferring transmission vary widely even for one pathogen. As an alternative to reliance on a strict SNP threshold, we suggest that the key inferential target when studying the spread of an infectious disease is the number of transmission events separating cases. Here we describe a new framework for deciding whether two pathogen genomes should be considered as part of the same transmission cluster, based jointly on the number of SNP differences and the length of time over which those differences have accumulated. Our approach allows us to probabilistically characterize the number of inferred transmission events that separate cases. We show how this framework can be modified to consider variable mutation rates across the genome (e.g. SNPs associated with drug resistance) and we indicate how the methodology can be extended to incorporate epidemiological data such as spatial proximity. We use recent data collected from tuberculosis studies from British Columbia, Canada and the Republic of Moldova to apply and compare our clustering method to the SNP threshold approach. In the British Columbia data, different cases break off from the main clusters as cut-off thresholds are lowered; the transmission-based method obtains slightly different clusters than the SNP cut-offs. For the Moldova data, straightforward application of the methods shows no appreciable difference, but when we take into account the fact that resistance conferring sites likely do not follow the same mutation clock as most sites due to selection, the transmission-based approach differs from the SNP cut-off method. Outbreak simulations confirm that our transmission based method is at least as good at identifying direct transmissions as a SNP cut-off. We conclude that the new method is a promising step towards establishing a more robust identification of outbreaks.

genomics↗

bioRxiv2018

Trends in CRP, D-dimer and fibrinogen during therapy for HIV associated multidrug resistant tuberculosis

BackgroundHIV positive adults on treatment for multidrug-resistant tuberculosis (MDR-TB) experience high mortality. Biomarkers of HIV/MDR-TB treatment response may enable earlier treatment modifications that improve outcomes.\n\nMethodsTo determine whether trends in C-reactive protein (CRP), D-dimer and fibrinogen predict treatment outcome among those with HIV/MDR-TB co-infection we studied 20 HIV positive participants initiating therapy for MDR-TB. Serum CRP, fibrinogen, and D-dimer were measured at baseline and serially while on treatment. Results: At baseline, all biomarkers were elevated with median CRP 86.15 mg/L (IQR 29.25-149.32), D-dimer 0.85 g/mL (IQR 0.34-1.80) and fibrinogen 4.11 g/L (IQR 3.75-6.31). CRP decreased significantly within 10 days of treatment initiation and fibrinogen within 28 days; D-dimer did not change significantly. 5 (25%) participants died. Older age (median age of 38y among survivors and 54y among deceased, p=0.008) and higher baseline fibrinogen (3.86 g/L among survivors and 6.37 g/L among deceased, p=0.02) were significantly associated with death. Higher CRP concentrations at the beginning of each measurement interval were significantly associated with a higher risk of death during that interval.\n\nConclusionTrends in fibrinogen and CRP may be useful for evaluating early response to treatment among individuals with HIV/MDR-TB co-infection.

immunology↗