StackHCV: a web-based integrative machine-learning framework for large-scale identification of hepatitis C virus NS5B inhibitors

Background Reducing the burden of the hepatitis C virus (HCV) requires large-scale deployment of intervention programmes, which can be informed by the dynamic pattern of HCV spread. In Spain, ongoing transmission of HCV is mostly fuelled by people who inject drugs (PWID) infected with subtype 1a (HCV1a). Aim Our aim was to map how infections spread within and between populations, which could help formulate more effective intervention programmes to halt the HCV1a epidemic in Spain. Methods Epidemiological links between HCV1a viruses from a convenience sample of 283 patients in Spain, mostly PWID, collected between 2014 and 2016, and 1,317, 1,291 and 1,009 samples collected abroad between 1989 and 2016 were reconstructed using sequences covering the NS3, NS5A and NS5B genes. To efficiently do so, fast maximum likelihood-based tree estimation was coupled to a flexible Bayesian discrete phylogeographic inference method. Results The transmission network structure of the Spanish HCV1a epidemic was shaped by continuous seeding of HCV1a into Spain, almost exclusively from North America and European countries. The latter became increasingly relevant and have dominated in recent times. Export from Spain to other countries in Europe was also strongly supported, although Spain was a net sink for European HCV1a lineages. Spatial reconstructions showed that the epidemic in Spain is diffuse, without large, dominant within-country networks. Conclusion To boost the effectiveness of local intervention efforts, concerted supra-national strategies to control HCV1a transmission are needed, with a strong focus on the most important drivers of ongoing transmission, i.e. PWID and other high-risk populations.

Download Full-text

Machine learning models to predict disease progression among veterans with hepatitis C virus

PLoS ONE ◽

10.1371/journal.pone.0208141 ◽

2019 ◽

Vol 14 (1) ◽

pp. e0208141 ◽

Cited By ~ 9

Author(s):

Monica A. Konerman ◽

Lauren A. Beste ◽

Tony Van ◽

Boang Liu ◽

Xuefei Zhang ◽

...

Keyword(s):

Machine Learning ◽

Hepatitis C Virus ◽

Hepatitis C ◽

Disease Progression ◽

Learning Models ◽

Machine Learning Models ◽

Predict Disease Progression

Download Full-text

Hepatitis C Guidance 2018 Update: AASLD-IDSA Recommendations for Testing, Managing, and Treating Hepatitis C Virus Infection

Clinical Infectious Diseases ◽

10.1093/cid/ciy585 ◽

2018 ◽

Vol 67 (10) ◽

pp. 1477-1492 ◽

Cited By ~ 214

Author(s):

◽

Raymond T Chung ◽

Marc G Ghany ◽

Arthur Y Kim ◽

Kristen M Marks ◽

...

Keyword(s):

Hepatitis C Virus ◽

Hepatitis C ◽

American Association ◽

Antiviral Agents ◽

Therapeutic Agents ◽

Hepatitis C Virus Infection ◽

Web Based ◽

Direct Acting Antiviral ◽

Direct Acting ◽

Direct Acting Antiviral Agents

Abstract Recognizing the importance of timely guidance regarding the rapidly evolving field of hepatitis C management, the American Association for the Study of Liver Diseases (AASLD) and the Infectious Diseases Society of America (IDSA) developed a web-based process for the expeditious formulation and dissemination of evidence-based recommendations. Launched in 2014, the hepatitis C virus (HCV) guidance website undergoes periodic updates as necessitated by availability of new therapeutic agents and/or research data. A major update was released electronically in September 2017, prompted primarily by approval of new direct-acting antiviral agents and expansion of the guidance’s scope. This update summarizes the latest release of the HCV guidance and focuses on new or amended recommendations since the previous September 2015 print publication. The recommendations herein were developed by volunteer hepatology and infectious disease experts representing AASLD and IDSA and have been peer reviewed and approved by each society’s governing board.

Download Full-text

Stochastic semi-supervised learning to prioritise genes from high-throughput genomic screens

10.1101/655449 ◽

2019 ◽

Author(s):

Dimitrios Vitsios ◽

Slavé Petrovski

Keyword(s):

Machine Learning ◽

Supervised Learning ◽

Large Scale ◽

Association Studies ◽

Protein Coding ◽

Knowledge Resources ◽

Learning Framework ◽

Significant Enrichment ◽

Disease Associated Genes ◽

Disease Specific

AbstractAccess to large-scale genomics datasets has increased the utility of hypothesis-free genome-wide analyses that result in candidate lists of genes. Often these analyses highlight several gene signals that might contribute to pathogenesis but are insufficiently powered to reach experiment-wide significance. This often triggers a process of laborious evaluation of highly-ranked genes through manual inspection of various public knowledge resources to triage those considered sufficiently interesting for deeper investigation. Here, we introduce a novel multi-dimensional, multi-step machine learning framework to objectively and more holistically assess biological relevance of genes to disease studies, by relying on a plethora of gene-associated annotations. We developed mantis-ml to serve as an automated machine learning (AutoML) framework, following a stochastic semi-supervised learning approach to rank known and novel disease-associated genes through iterative training and prediction sessions of random balanced datasets across the protein-coding exome (n=18,626 genes). We applied this framework on a range of disease-specific areas and as a generic disease likelihood estimator, achieving an average Area Under Curve (AUC) prediction performance of 0.85. Critically, to demonstrate applied utility on exome-wide association studies, we overlapped mantis-ml disease-specific predictions with data from published cohort-level association studies. We retrieved statistically significant enrichment of high mantis-ml predictions among the top-ranked genes from hypothesis-free cohort-level statistics (p<0.05), suggesting the capture of true prioritisation signals. We believe that mantis-ml is a novel easy-to-use tool to support objectively triaging gene discovery and overall enhancing our understanding of complex genotype-phenotype associations.

Download Full-text

Impact of virus subtype and host IFNL4 genotype on large-scale RNA structure formation in the genome of hepatitis C virus

10.1101/2020.06.16.155150 ◽

2020 ◽

Cited By ~ 1

Author(s):

P. Simmonds ◽

L. Cuypers ◽

W.L. Irving ◽

J. McLauchlan ◽

G.S. Cooke ◽

...

Keyword(s):

Hepatitis C Virus ◽

Hepatitis C ◽

Rna Structure ◽

Large Scale ◽

Rna Virus ◽

Energy Difference ◽

Folding Energy ◽

Stem Loop ◽

Coding Regions ◽

Progressive Liver Disease

ABSTRACTMechanisms underlying the ability of hepatitis C virus (HCV) to establish persistent infections and induce progressive liver disease remain poorly understood. HCV is one of several positive-stranded RNA viruses capable of establishing persistence in their immunocompetent vertebrate hosts, an attribute associated with formation of large scale RNA structure in their genomic RNA. We developed novel methods to analyse and visualise genome-scale ordered RNA structure (GORS) predicted from the increasingly large datasets of complete genome sequences of HCV. Structurally conserved RNA secondary structure in coding regions of HCV localised exclusively to polyprotein ends (core, NS5B). Coding regions elsewhere were also intensely structured based on elevated minimum folding energy difference (MFED) values, but the actual stem-loop elements involved in genome folding were structurally entirely distinct, even between subtypes 1a and 1b. Dynamic remodelling was further evident from comparison of HCV strains in different host genetic background. Significantly higher MFED values, greater suppression of UpA dinucleotide frequencies and restricted diversification were found in subjects with the TT genotype of the rs12979860 SNP in the IFNL4 gene compared to the CC (non-expressing) allele. These structural and compositional associations with expression of interferon-λ4 were recapitulated on a larger scale by higher MFED values and greater UpA suppression of genotype 1 compared to genotype 3a, associated with previously reported HCV genotype-associated differences in hepatic interferon-stimulated gene induction. Associations between innate cellular responses with HCV structure and further evolutionary constraints represents an important new element in RNA virus evolution and the adaptive interplay between virus and host.

Download Full-text

Impact of virus subtype and host IFNL4 genotype on large-scale RNA structure formation in the genome of hepatitis C virus

RNA ◽

10.1261/rna.075465.120 ◽

2020 ◽

Vol 26 (11) ◽

pp. 1541-1556 ◽

Cited By ~ 1

Author(s):

Peter Simmonds ◽

Lize Cuypers ◽

Will L. Irving ◽

John McLauchlan ◽

Graham S. Cooke ◽

...

Keyword(s):

Hepatitis C Virus ◽

Hepatitis C ◽

Structure Formation ◽

Rna Structure ◽

Large Scale ◽

Virus Subtype

Download Full-text

Rational Design to Large-Scale Synthesis: Development of GSK8175 for the Treatment of Hepatitis C Virus Infection

ACS Symposium Series - Complete Accounts of Integrated Drug Discovery and Development: Recent Examples from the Pharmaceutical Industry Volume 2 ◽

10.1021/bk-2019-1332.ch011 ◽

2019 ◽

pp. 297-322

Author(s):

Andrew J. Peat ◽

Shiping Xie

Keyword(s):

Hepatitis C Virus ◽

Hepatitis C ◽

Virus Infection ◽

Rational Design ◽

Large Scale ◽

Hepatitis C Virus Infection ◽

Large Scale Synthesis

Download Full-text

Large-scale structure-activity relationship study of hepatitis C virus NS5B polymerase inhibition using SMILES-based descriptors

Molecular Diversity ◽

10.1007/s11030-015-9614-2 ◽

2015 ◽

Vol 19 (4) ◽

pp. 955-964 ◽

Cited By ~ 10

Author(s):

Apilak Worachartcheewan ◽

Virapong Prachayasittikul ◽

Alla P. Toropova ◽

Andrey A. Toropov ◽

Chanin Nantasenamat

Keyword(s):

Hepatitis C Virus ◽

Hepatitis C ◽

Large Scale ◽

Large Scale Structure ◽

Structure Activity Relationship ◽

Scale Structure ◽

Activity Relationship ◽

Structure Activity ◽

Relationship Study ◽

Ns5b Polymerase

Download Full-text

Unifying machine learning and quantum chemistry with a deep neural network for molecular wavefunctions

Nature Communications ◽

10.1038/s41467-019-12875-2 ◽

2019 ◽

Vol 10 (1) ◽

Cited By ~ 64

Author(s):

K. T. Schütt ◽

M. Gastegger ◽

A. Tkatchenko ◽

K.-R. Müller ◽

R. J. Maurer

Keyword(s):

Machine Learning ◽

Quantum Chemistry ◽

Degrees Of Freedom ◽

Large Scale ◽

Materials Science ◽

Chemical Space ◽

Chemical Properties ◽

Molecular Structures ◽

Learning Framework ◽

Molecular Wavefunctions

AbstractMachine learning advances chemistry and materials science by enabling large-scale exploration of chemical space based on quantum chemical calculations. While these models supply fast and accurate predictions of atomistic chemical properties, they do not explicitly capture the electronic degrees of freedom of a molecule, which limits their applicability for reactive chemistry and chemical analysis. Here we present a deep learning framework for the prediction of the quantum mechanical wavefunction in a local basis of atomic orbitals from which all other ground-state properties can be derived. This approach retains full access to the electronic structure via the wavefunction at force-field-like efficiency and captures quantum mechanics in an analytically differentiable representation. On several examples, we demonstrate that this opens promising avenues to perform inverse design of molecular structures for targeting electronic property optimisation and a clear path towards increased synergy of machine learning and quantum chemistry.

Download Full-text