scholarly journals Developing anin silicominimum inhibitory concentration panel test forKlebsiella pneumoniae

2017 ◽  
Author(s):  
Marcus Nguyen ◽  
Thomas Brettin ◽  
S. Wesley Long ◽  
Randall J. Olsen ◽  
James M. Musser ◽  
...  
Keyword(s):  
Machine Learning ◽  
Pathogenic Bacteria ◽  
Prediction Models ◽  
Sequence Data ◽  
Whole Genome Sequence ◽  
Clinical Test ◽  
Whole Genome ◽  
Conventional Culture ◽  
Machine Learning Model ◽  
Antimicrobial Resistance Genes

ABSTRACTAntimicrobial resistant infections are a serious public health threat worldwide. Whole genome sequencing approaches to apidly identify pathogens and predict antibiotic resistance phenotypes are becoming more feasible and may offer a way to reduce clinical test turnaround times compared to conventional culture-based methods, and in turn, improve patient outcomes. In this study, we use whole genome sequence data from 1668 clinical isolates ofKlebsiella pneumoniaeto develop a XGBoost-based machine learning model that accurately predicts minimum inhibitory concentrations (MICs) for 20 antibiotics. The overall accuracy of the model, within ±1 two-fold dilution factor, is 92%. Individual accuracies are≥90% for 15/20 antibiotics. We show that the MICs predicted by the model correlate with known antimicrobial resistance genes. Importantly, the genome-wide approach described in this study offers a way to predict MICs for isolates without knowledge of the underlying gene content. This study shows that machine learning can be used to build a completein silicoMIC prediction panel forK. pneumoniaeand provides a framework for building MIC prediction models for other pathogenic bacteria.

scholarly journals Using machine learning to predict antimicrobial minimum inhibitory concentrations and associated genomic features for nontyphoidalSalmonella

2018 ◽  
Author(s):  
Marcus Nguyen ◽  
S. Wesley Long ◽  
Patrick F. McDermott ◽  
Randall J. Olsen ◽  
Robert Olson ◽  
...  
Keyword(s):  
United States ◽  
Machine Learning ◽  
Prediction Models ◽  
The United States ◽  
Clinical Diagnostics ◽  
Gene Content ◽  
Whole Genome Sequence ◽  
Whole Genome ◽  
Minimum Inhibitory Concentrations ◽  
Antimicrobial Resistance Gene

NontyphoidalSalmonellaspecies are the leading bacterial cause of food-borne disease in the United States. Whole genome sequences and paired antimicrobial susceptibility data are available forSalmonellastrains because of surveillance efforts from public health agencies. In this study, a collection of 5,278 nontyphoidalSalmonellagenomes, collected over 15 years in the United States, were used to generate XGBoost-based machine learning models for predicting minimum inhibitory concentrations (MICs) for 15 antibiotics. The MIC prediction models have average accuracies between 95-96% within ± 1 two-fold dilution factor and can predict MICs with noa prioriinformation about the underlying gene content or resistance phenotypes of the strains. By selecting diverse genomes for training sets, we show that highly accurate MIC prediction models can be generated with fewer than 500 genomes. We also show that our approach for predicting MICs is stable over time despite annual fluctuations in antimicrobial resistance gene content in the sampled genomes. Finally, using feature selection, we explore the important genomic regions identified by the models for predicting MICs. To date, this is one of the largest MIC modeling studies to be published. Our strategy for developing whole genome sequence-based models for surveillance and clinical diagnostics can be readily applied to other important human pathogens.

scholarly journals Evaluation of parameters affecting performance and reliability of machine learning-based antibiotic susceptibility testing from whole genome sequencing data

2019 ◽  
Author(s):  
Allison L. Hicks ◽  
Nicole Wheeler ◽  
Leonor Sánchez-Busó ◽  
Jennifer L. Rakeman ◽  
Simon R. Harris ◽  
...  
Keyword(s):  
Machine Learning ◽  
Antibiotic Resistance ◽  
Antibiotic Susceptibility ◽  
Sequence Data ◽  
Model Performance ◽  
Outcome Data ◽  
Whole Genome Sequence ◽  
Whole Genome Sequencing Data ◽  
Whole Genome ◽  
Sequencing Data

AbstractPrediction of antibiotic resistance phenotypes from whole genome sequencing data by machine learning methods has been proposed as a promising platform for the development of sequence-based diagnostics. However, there has been no systematic evaluation of factors that may influence performance of such models, how they might apply to and vary across clinical populations, and what the implications might be in the clinical setting. Here, we performed a meta-analysis of seven large Neisseria gonorrhoeae datasets, as well as Klebsiella pneumoniae and Acinetobacter baumannii datasets, with whole genome sequence data and antibiotic susceptibility phenotypes using set covering machine classification, random forest classification, and random forest regression models to predict resistance phenotypes from genotype. We demonstrate how model performance varies by drug, dataset, resistance metric, and species, reflecting the complexities of generating clinically relevant conclusions from machine learning-derived models. Our findings underscore the importance of incorporating relevant biological and epidemiological knowledge into model design and assessment and suggest that doing so can inform tailored modeling for individual drugs, pathogens, and clinical populations. We further suggest that continued comprehensive sampling and incorporation of up-to-date whole genome sequence data, resistance phenotypes, and treatment outcome data into model training will be crucial to the clinical utility and sustainability of machine learning-based molecular diagnostics.Author SummaryMachine learning-based prediction of antibiotic resistance from bacterial genome sequences represents a promising tool to rapidly determine the antibiotic susceptibility profile of clinical isolates and reduce the morbidity and mortality resulting from inappropriate and ineffective treatment. However, while there has been much focus on demonstrating the diagnostic potential of these modeling approaches, there has been little assessment of potential caveats and prerequisites associated with implementing predictive models of drug resistance in the clinical setting. Our results highlight significant biological and technical challenges facing the application of machine learning-based prediction of antibiotic resistance as a diagnostic tool. By outlining specific factors affecting model performance, our findings provide a framework for future work on modeling drug resistance and underscore the necessity of continued comprehensive sampling and reporting of treatment outcome data for building reliable and sustainable diagnostics.

TIGER: inferring DNA replication timing from whole-genome sequence data

2021 ◽  
Author(s):  
Amnon Koren ◽  
Dashiell J Massey ◽  
Alexa N Bracci
Keyword(s):  
Dna Replication ◽  
Genome Sequence ◽  
Genomic Dna ◽  
Sequence Data ◽  
Replication Timing ◽  
Whole Genome Sequence ◽  
Supplementary Information ◽  
Whole Genome ◽  
Genome Sequence Data ◽  
Dna Replication Timing

Abstract Motivation Genomic DNA replicates according to a reproducible spatiotemporal program, with some loci replicating early in S phase while others replicate late. Despite being a central cellular process, DNA replication timing studies have been limited in scale due to technical challenges. Results We present TIGER (Timing Inferred from Genome Replication), a computational approach for extracting DNA replication timing information from whole genome sequence data obtained from proliferating cell samples. The presence of replicating cells in a biological specimen leads to non-uniform representation of genomic DNA that depends on the timing of replication of different genomic loci. Replication dynamics can hence be observed in genome sequence data by analyzing DNA copy number along chromosomes while accounting for other sources of sequence coverage variation. TIGER is applicable to any species with a contiguous genome assembly and rivals the quality of experimental measurements of DNA replication timing. It provides a straightforward approach for measuring replication timing and can readily be applied at scale. Availability and Implementation TIGER is available at https://github.com/TheKorenLab/TIGER. Supplementary information Supplementary data are available at Bioinformatics online

scholarly journals Whole genome sequence data of Mycobacterium tuberculosis XDR strain, isolated from patient in Kazakhstan

Data in Brief ◽  
2020 ◽  
Vol 33 ◽  
pp. 106416
Author(s):  
Asset Daniyarov ◽  
Askhat Molkenov ◽  
Saule Rakhimova ◽  
Ainur Akhmetova ◽  
Zhannur Nurkina ◽  
...  
Keyword(s):  
Mycobacterium Tuberculosis ◽  
Genome Sequence ◽  
Sequence Data ◽  
Whole Genome Sequence ◽  
Whole Genome ◽  
Genome Sequence Data

scholarly journals Elucidating the genetic basis of an oligogenic birth defect using whole genome sequence data in a non-model organism, Bubalus bubalis

2017 ◽  
Vol 7 (1) ◽  
Author(s):  
Lynsey K. Whitacre ◽  
Jesse L. Hoff ◽  
Robert D. Schnabel ◽  
Sara Albarella ◽  
Francesca Ciotola ◽  
...  
Keyword(s):  
Genome Sequence ◽  
Birth Defect ◽  
Genetic Basis ◽  
Sequence Data ◽  
Model Organism ◽  
Bubalus Bubalis ◽  
Whole Genome Sequence ◽  
Whole Genome ◽  
Genome Sequence Data

scholarly journals Whole genome characterization of strains belonging to the Ralstonia solanacearum species complex and in silico analysis of TaqMan assays for detection in this heterogenous species complex

2021 ◽  
Author(s):  
Viola Kurm ◽  
Ilse Houwers ◽  
Claudia E. Coipan ◽  
Peter Bonants ◽  
Cees Waalwijk ◽  
...  
Keyword(s):  
Ralstonia Solanacearum ◽  
In Silico ◽  
Species Complex ◽  
Sequence Data ◽  
In Silico Analysis ◽  
Whole Genome Sequence ◽  
Whole Genome ◽  
Genome Sequences ◽  
Pcr Assays

AbstractIdentification and classification of members of the Ralstonia solanacearum species complex (RSSC) is challenging due to the heterogeneity of this complex. Whole genome sequence data of 225 strains were used to classify strains based on average nucleotide identity (ANI) and multilocus sequence analysis (MLSA). Based on the ANI score (>95%), 191 out of 192(99.5%) RSSC strains could be grouped into the three species R. solanacearum, R. pseudosolanacearum, and R. syzygii, and into the four phylotypes within the RSSC (I,II, III, and IV). R. solanacearum phylotype II could be split in two groups (IIA and IIB), from which IIB clustered in three subgroups (IIBa, IIBb and IIBc). This division by ANI was in accordance with MLSA. The IIB subgroups found by ANI and MLSA also differed in the number of SNPs in the primer and probe sites of various assays. An in-silico analysis of eight TaqMan and 11 conventional PCR assays was performed using the whole genome sequences. Based on this analysis several cases of potential false positives or false negatives can be expected upon the use of these assays for their intended target organisms. Two TaqMan assays and two PCR assays targeting the 16S rDNA sequence should be able to detect all phylotypes of the RSSC. We conclude that the increasing availability of whole genome sequences is not only useful for classification of strains, but also shows potential for selection and evaluation of clade specific nucleic acid-based amplification methods within the RSSC.

46 Footprints of Selection in Angus and Hanwoo Beef Cattle Using Imputed Whole Genome Sequence Data

2021 ◽  
Vol 99 (Supplement_3) ◽  
pp. 25-25
Author(s):  
Muhammad Yasir Nawaz ◽  
Rodrigo Pelicioni Savegnago ◽  
Cedric Gondro
Keyword(s):  
Beef Cattle ◽  
Genome Sequence ◽  
Sequence Data ◽  
Whole Genome Sequence ◽  
Fixation Index ◽  
Whole Genome ◽  
Extended Haplotype Homozygosity ◽  
Extended Haplotype ◽  
Genome Sequence Data ◽  
Genomic Regions

Abstract In this study, we detected genome wide footprints of selection in Hanwoo and Angus beef cattle using different allele frequency and haplotype-based methods based on imputed whole genome sequence data. Our dataset included 13,202 Angus and 10,437 Hanwoo animals with 10,057,633 and 13,241,550 imputed SNPs, respectively. A subset of data with 6,873,624 common SNPs between the two populations was used to estimate signatures of selection parameters, both within (runs of homozygosity and extended haplotype homozygosity) and between (allele fixation index, extended haplotype homozygosity) the breeds in order to infer evidence of selection. We observed that correlations between various measures of selection ranged between 0.01 to 0.42. Assuming these parameters were complementary to each other, we combined them into a composite selection signal to identify regions under selection in both beef breeds. The composite signal was based on the average of fractional ranks of individual selection measures for every SNP. We identified some selection signatures that were common between the breeds while others were independent. We also observed that more genomic regions were selected in Angus as compared to Hanwoo. Candidate genes within significant genomic regions may help explain mechanisms of adaptation, domestication history and loci for important traits in Angus and Hanwoo cattle. In the future, we will use the top SNPs under selection for genomic prediction of carcass traits in both breeds.

Sign in / Sign up

Export Citation Format

Share Document