Machine-Learning Model for Prediction of Cefepime Susceptibility in Escherichia coli from Whole-Genome Sequencing Data.

The declining cost of performing bacterial whole-genome sequencing (WGS) coupled with the availability of large libraries of sequence data for well-characterized isolates have enabled the application of machine-learning (ML) methods to the development of nonlinear sequence-based predictive models. We tested the ML-based model developed by Next Gen Diagnostics for prediction of cefepime phenotypic susceptibility results in Escherichia coli. A cohort of 100 isolates of E. coli recovered from urine (n = 77) and blood (n = 23) cultures were used. The cefepime MIC was determined in triplicate by reference broth microdilution and classified as susceptible (MIC of ≤2 µg/mL) or not susceptible (MIC of ≥4 µg/mL) using the 2022 Clinical and Laboratory Standards Institute breakpoints. Five isolates generated both susceptible and not susceptible MIC results, yielding categorical agreement of 95% for the reference method to itself. Categorical agreement of ML to MIC interpretations was 97%, with 2 very major (false, susceptible) and 1 major (false, not susceptible) errors. One very major error occurred for an isolate with blaCTX-M-27 (MIC mode, ≥32 µg/mL) and one for an isolate with blaTEM-34 for which the MIC cefepime mode was 4 µg/mL. One major error was for an isolate with blaCTX-M-27 but with a MIC mode of 2 µg/mL. These preliminary data demonstrated performance of ML for a clinically important antimicrobial-species pair at a caliber similar to phenotypic methods, encouraging wider development of sequence-based susceptibility prediction and its validation and use in clinical practice.

Large-Scale Evaluation of a Rapid Fully Automated Analysis Platform to Detect and Refute Outbreaks Based on MRSA Genome Comparisons.

Genomic epidemiology of methicillin-resistant Staphylococcus aureus (MRSA) could transform outbreak investigations, but its clinical introduction is hampered by the lack of automated data analysis tools to rapidly and accurately define transmission based on sequence relatedness. We aimed to evaluate a fully automated bioinformatics system for MRSA genome analysis versus a bespoke researcher-led manual informatics pipeline. We analyzed 781 MRSA genomes from 777 consecutive patients identified over a 9-month period in a clinical microbiology laboratory in the United Kingdom. Outputs were bacterial species identification, detection of mec genes, assignment to sequence types (STs), identification of pairwise relatedness using a definition of ≤25 single nucleotide polymorphisms (SNPs) apart, and use of genetic relatedness to identify clusters. There was full concordance between the two analysis methods for species identification, detection of mec genes, and ST assignment. A total of 3,311 isolate pairs ≤25 SNPs apart were identified by at least one method. These had a median (range) SNP difference between the two methods of 1.2 SNPs (0 to 22 SNPs), with most isolate pairs (87%) varying by ≤2 SNPs. This similarity increased when the research pipeline was modified to use a clonal-complex-specific reference (median 0 SNP difference, 91% varying by ≤2 SNPs). Both pipelines clustered 338 isolates/334 patients into 66 unique clusters based on genetic relatedness. We conclude that the automated transmission detection tool worked at least as well as a researcher-led manual analysis and indicates how such tools could support the rapid use of MRSA genomic epidemiology in infection control practice. IMPORTANCE It has been clearly established that genome sequencing of MRSA improves the accuracy of health care-associated outbreak investigations, including the confirmation and exclusion of outbreaks and identification of patients involved. This could lead to more targeted infection control actions but its use in clinical practice is prevented by several barriers, one of which is the availability of genome analysis tools that do not depend on specialist knowledge to analyze or interpret the results. We evaluated a prototype of a fully automated bioinformatics system for MRSA genome analysis versus a bespoke researcher-led manual informatics pipeline, using genomes from 777 patients over a period of 9 months. The performance was at least equivalent to the researcher-led manual genomic analysis. This indicates the feasibility of automated analysis and represents one more step toward the routine use of pathogen sequencing in infection prevention and control practice.

Defining metrics for whole-genome sequence analysis of MRSA in clinical practice.

Bacterial sequencing will become increasingly adopted in routine microbiology laboratories. Here, we report the findings of a technical evaluation of almost 800 clinical methicillin-resistant Staphylococcus aureus (MRSA) isolates, in which we sought to define key quality metrics to support MRSA sequencing in clinical practice. We evaluated the accuracy of mapping to a generic reference versus clonal complex (CC)-specific mapping, which is more computationally challenging. Focusing on isolates that were genetically related (<50 single nucleotide polymorphisms (SNPs)) and belonged to prevalent sequence types, concordance between these methods was 99.5 %. We use MRSA MPROS0386 to control for base calling accuracy by the sequencer, and used multiple repeat sequences of the control to define a permitted range of SNPs different to the mapping reference for this control (equating to 3 standard deviations from the mean). Repeat sequences of the control were also used to demonstrate that SNP calling was most accurate across differing coverage depths (above 35×, the lowest depth in our study) when the depth required to call a SNP as present was at least 4-8×. Using 786 MRSA sequences, we defined a robust measure for mec gene detection to reduce false-positives arising from contamination, which was no greater than 2 standard deviations below the average depth of coverage across the genome. Sequencing from bacteria harvested from clinical plates runs an increased risk of contamination with the same or different species, and we defined a cut-off of 30 heterozygous sites >50 bp apart to identify same-species contamination for MRSA. These metrics were combined into a quality-control (QC) flowchart to determine whether sequence runs and individual clinical isolates passed QC, which could be adapted by future automated analysis systems to enable rapid hands-off sequence analysis by clinical laboratories.

Evaluation of a fully automated bioinformatics tool to predict antibiotic resistance from MRSA genomes.

OBJECTIVES: The genetic prediction of phenotypic antibiotic resistance based on analysis of WGS data is becoming increasingly feasible, but a major barrier to its introduction into routine use is the lack of fully automated interpretation tools. Here, we report the findings of a large evaluation of the Next Gen Diagnostics (NGD) automated bioinformatics analysis tool to predict the phenotypic resistance of MRSA. METHODS: MRSA-positive patients were identified in a clinical microbiology laboratory in England between January and November 2018. One MRSA isolate per patient together with all blood culture isolates (total n = 778) were sequenced on the Illumina MiniSeq instrument in batches of 21 clinical MRSA isolates and three controls. RESULTS: The NGD system activated post-sequencing and processed the sequences to determine susceptible/resistant predictions for 11 antibiotics, taking around 11 minutes to analyse 24 isolates sequenced on a single sequencing run. NGD results were compared with phenotypic susceptibility testing performed by the clinical laboratory using the disc diffusion method and EUCAST breakpoints. Following retesting of discrepant results, concordance between phenotypic results and NGD genetic predictions was 99.69%. Further investigation of 22 isolate genomes associated with persistent discrepancies revealed a range of reasons in 12 cases, but no cause could be found for the remainder. Genetic predictions generated by the NGD tool were compared with predictions generated by an independent research-based informatics approach, which demonstrated an overall concordance between the two methods of 99.97%. CONCLUSIONS: We conclude that the NGD system provides rapid and accurate prediction of the antibiotic susceptibility of MRSA.

Pilot Evaluation of a Fully Automated Bioinformatics System for Analysis of Methicillin-Resistant Staphylococcus aureus Genomes and Detection of Outbreaks.

Genomic surveillance that combines bacterial sequencing and epidemiological information will become the gold standard for outbreak detection, but its clinical translation is hampered by the lack of automated interpretation tools. We performed a prospective pilot study to evaluate the analysis of methicillin-resistant Staphylococcus aureus (MRSA) genomes using the Next Gen Diagnostics (NGD) automated bioinformatics system. Seventeen unselected MRSA-positive patients were identified in a clinical microbiology laboratory in England over a period of 2 weeks in 2018, and 1 MRSA isolate per case was sequenced on the Illumina MiniSeq instrument. The NGD system automatically activated after sequencing and processed fastq folders to determine species, multilocus sequence type, the presence of a mec gene, antibiotic susceptibility predictions, and genetic relatedness based on mapping to a reference MRSA genome and detection of pairwise core genome single-nucleotide polymorphisms. The NGD system required 90 s per sample to automatically analyze data from each run, the results of which were automatically displayed. The same data were independently analyzed using a research-based approach. There was full concordance between the two analysis methods regarding species (S. aureus), detection of mecA, sequence type assignment, and detection of genetic determinants of resistance. Both analysis methods identified two MRSA clusters based on relatedness, one of which contained 3 cases that were involved in an outbreak linked to a clinic and ward associated with diabetic patient care. We conclude that, in this pilot study, the NGD system provided rapid and accurate data that could support infection control practices.

Attitudes of nearly 7000 health professionals, genomic researchers and publics toward the return of incidental results from sequencing research.

Genome-wide sequencing in a research setting has the potential to reveal health-related information of personal or clinical utility for the study participant. There is increasing pressure to return research findings to participants that may not be related to the project aims, particularly when these could be used to prevent disease. Such secondary, unsolicited or 'incidental findings' (IFs) may be discovered unintentionally when interpreting sequence data, or as the result of a deliberate opportunistic screen. This cross-sectional, web-based survey investigated attitudes of 6944 individuals from 75 countries towards returning IFs from genome research. Participants included four relevant stakeholder groups: 4961 members of the public, 533 genetic health professionals, 843 non-genetic health professionals and 607 genomic researchers who were invited via traditional media, social media and professional e-mail list-serve. Treatability and perceived utility of incidental results were deemed important with 98% of stakeholders personally interested in learning about preventable life-threatening conditions. Although there was a generic interest in receiving genomic information, stakeholders did not expect researchers to opportunistically screen for IFs in a research setting. On many items, genetic health professionals had significantly more conservative views compared with other stakeholders. This finding demonstrates a disconnect between the views of those handling the findings of research and those participating in research. Exploring, evaluating and ultimately addressing this disconnect should form a priority for researchers and clinicians alike. This social sciences study offers the largest dataset, published to date, of attitudes towards issues surrounding the return of IFs from sequencing research.

Potential research participants support the return of raw sequence data.

Health-related results that are discovered in the process of genomic research should only be returned to research participants after being clinically validated and then delivered and followed up within a health service. Returning such results may be difficult for genomic researchers who are limited by resources or unable to access appropriate clinicians. Raw sequence data could, in theory, be returned instead. This might appear nonsensical as, on its own, it is a meaningless code with no clinical value. Yet, as and when direct to consumer genomics services become more widely available (and can be endorsed by independent health professionals and genomic researchers alike), the return of such data could become a realistic proposition. We explore attitudes from <7000 members of the public, genomic researchers, genetic health professionals and non-genetic health professionals and ask participants to suggest what they would do with a raw sequence, if offered it. Results show 62% participants were interested in using it to seek out their own clinical interpretation. Whilst we do not propose that raw sequence data should be returned at the moment, we suggest that should this become feasible in the future, participants of sequencing studies may possibly support this.

Genetic diagnosis of developmental disorders in the DDD study: a scalable analysis of genome-wide research data.

BACKGROUND: Human genome sequencing has transformed our understanding of genomic variation and its relevance to health and disease, and is now starting to enter clinical practice for the diagnosis of rare diseases. The question of whether and how some categories of genomic findings should be shared with individual research participants is currently a topic of international debate, and development of robust analytical workflows to identify and communicate clinically relevant variants is paramount. METHODS: The Deciphering Developmental Disorders (DDD) study has developed a UK-wide patient recruitment network involving over 180 clinicians across all 24 regional genetics services, and has performed genome-wide microarray and whole exome sequencing on children with undiagnosed developmental disorders and their parents. After data analysis, pertinent genomic variants were returned to individual research participants via their local clinical genetics team. FINDINGS: Around 80,000 genomic variants were identified from exome sequencing and microarray analysis in each individual, of which on average 400 were rare and predicted to be protein altering. By focusing only on de novo and segregating variants in known developmental disorder genes, we achieved a diagnostic yield of 27% among 1133 previously investigated yet undiagnosed children with developmental disorders, whilst minimising incidental findings. In families with developmentally normal parents, whole exome sequencing of the child and both parents resulted in a 10-fold reduction in the number of potential causal variants that needed clinical evaluation compared to sequencing only the child. Most diagnostic variants identified in known genes were novel and not present in current databases of known disease variation. INTERPRETATION: Implementation of a robust translational genomics workflow is achievable within a large-scale rare disease research study to allow feedback of potentially diagnostic findings to clinicians and research participants. Systematic recording of relevant clinical data, curation of a gene-phenotype knowledge base, and development of clinical decision support software are needed in addition to automated exclusion of almost all variants, which is crucial for scalable prioritisation and review of possible diagnostic variants. However, the resource requirements of development and maintenance of a clinical reporting system within a research setting are substantial. FUNDING: Health Innovation Challenge Fund, a parallel funding partnership between the Wellcome Trust and the UK Department of Health.

DECIPHER: database for the interpretation of phenotype-linked plausibly pathogenic sequence and copy-number variation.

The DECIPHER database (https://decipher.sanger.ac.uk/) is an accessible online repository of genetic variation with associated phenotypes that facilitates the identification and interpretation of pathogenic genetic variation in patients with rare disorders. Contributing to DECIPHER is an international consortium of >200 academic clinical centres of genetic medicine and ≥1600 clinical geneticists and diagnostic laboratory scientists. Information integrated from a variety of bioinformatics resources, coupled with visualization tools, provides a comprehensive set of tools to identify other patients with similar genotype-phenotype characteristics and highlights potentially pathogenic genes. In a significant development, we have extended DECIPHER from a database of just copy-number variants to allow upload, annotation and analysis of sequence variants such as single nucleotide variants (SNVs) and InDels. Other notable developments in DECIPHER include a purpose-built, customizable and interactive genome browser to aid combined visualization and interpretation of sequence and copy-number variation against informative datasets of pathogenic and population variation. We have also introduced several new features to our deposition and analysis interface. This article provides an update to the DECIPHER database, an earlier instance of which has been described elsewhere [Swaminathan et al. (2012) DECIPHER: web-based, community resource for clinical interpretation of rare variants in developmental disorders. Hum. Mol. Genet., 21, R37-R44].

DECIPHER: web-based, community resource for clinical interpretation of rare variants in developmental disorders.

Patients with developmental disorders often harbour sub-microscopic deletions or duplications that lead to a disruption of normal gene expression or perturbation in the copy number of dosage-sensitive genes. Clinical interpretation for such patients in isolation is hindered by the rarity and novelty of such disorders. The DECIPHER project (https://decipher.sanger.ac.uk) was established in 2004 as an accessible online repository of genomic and associated phenotypic data with the primary goal of aiding the clinical interpretation of rare copy-number variants (CNVs). DECIPHER integrates information from a variety of bioinformatics resources and uses visualization tools to identify potential disease genes within a CNV. A two-tier access system permits clinicians and clinical scientists to maintain confidential linked anonymous records of phenotypes and CNVs for their patients that, with informed consent, can subsequently be shared with the wider clinical genetics and research communities. Advances in next-generation sequencing technologies are making it practical and affordable to sequence the whole exome/genome of patients who display features suggestive of a genetic disorder. This approach enables the identification of smaller intragenic mutations including single-nucleotide variants that are not accessible even with high-resolution genomic array analysis. This article briefly summarizes the current status and achievements of the DECIPHER project and looks ahead to the opportunities and challenges of jointly analysing structural and sequence variation in the human genome.

Interpretation of genomic copy number variants using DECIPHER.

Many patients suffering from developmental disorders have submicroscopic deletions or duplications affecting the copy number of dosage-sensitive genes or disrupting normal gene expression. Many of these changes are novel or extremely rare, making clinical interpretation problematic and genotype/phenotype correlations difficult. Identification of patients sharing a genomic rearrangement and having phenotypes in common increases certainty in the diagnosis and allows characterization of new syndromes. The DECIPHER database is an online repository of genotype and phenotype data whose chief objective is to facilitate the association of genomic variation with phenotype to enable the clinical interpretation of copy number variation (CNV). This unit shows how DECIPHER can be used to (1) search for consented patients sharing a defined chromosomal location, (2) navigate regions of interest using in-house visualization tools and the Ensembl genome browser, (3) analyze affected genes and prioritize them according to their likelihood of haploinsufficiency, (4) upload patient aberrations and phenotypes, and (5) create printouts at different levels of detail. By following this protocol, clinicians and researchers alike will be able to learn how to characterize their patients' chromosomal imbalances using DECIPHER.

Using caching and optimization techniques to improve performance of the Ensembl website.

BACKGROUND: The Ensembl web site has provided access to genomic information for almost 10 years. During this time the amount of data available through Ensembl has grown dramatically. At the same time, the World Wide Web itself has become a dramatically more important component of the scientific workflow and the way that scientists share and access data and scientific information. Since 2000, the Ensembl web interface has had three major updates and numerous smaller updates. These have largely been in response to expanding data types and valuable representations of existing data types. In 2007 it was realised that a radical new approach would be required in order to serve the project's future requirements, and development therefore focused on identifying suitable web technologies for implementation in the 2008 site redesign. RESULTS: By comparing the Ensembl website to well-known "Web 2.0" sites, we were able to identify two main areas in which cutting-edge technologies could be advantageously deployed: server efficiency and interface latency. We then evaluated the performance of the existing site using browser-based tools and Apache benchmarking, and selected appropriate technologies to overcome any issues found. Solutions included optimization of the Apache web server, introduction of caching technologies and widespread implementation of AJAX code. These improvements were successfully deployed on the Ensembl website in late 2008 and early 2009. CONCLUSIONS: Web 2.0 technologies provide a flexible and efficient way to access the terabytes of data now available from Ensembl, enhancing the user experience through improved website responsiveness and a rich, interactive interface.

Ensembl's 10th year.

Ensembl (http://www.ensembl.org) integrates genomic information for a comprehensive set of chordate genomes with a particular focus on resources for human, mouse, rat, zebrafish and other high-value sequenced genomes. We provide complete gene annotations for all supported species in addition to specific resources that target genome variation, function and evolution. Ensembl data is accessible in a variety of formats including via our genome browser, API and BioMart. This year marks the tenth anniversary of Ensembl and in that time the project has grown with advances in genome technology. As of release 56 (September 2009), Ensembl supports 51 species including marmoset, pig, zebra finch, lizard, gorilla and wallaby, which were added in the past year. Major additions and improvements to Ensembl since our previous report include the incorporation of the human GRCh37 assembly, enhanced visualisation and data-mining options for the Ensembl regulatory features and continued development of our software infrastructure.