Trainable high resolution melt curve machine learning classifier for large-scale reliable genotyping of sequence variants

Enhancing SARS-CoV-2 Lineage Surveillance through the Integration of a Simple and Direct qPCR-Based Protocol Adaptation with Established Machine Learning Algorithms

Emerging and evolving Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) lineages, adapted to changing epidemiological conditions, present unprecedented challenges to global public health systems. Here, we introduce an adapted analytical approach that complements genomic sequencing, applying a cost-effective quantitative polymerase chain reaction (qPCR)-based assay. Viral RNA samples from SARS-CoV-2 positive cases detected by diagnostic laboratories or public health network units in Ceará, Brazil, were tracked for genomic surveillance and analyzed by using paired-end sequencing combined with integrative genomic analysis. Validation of a key structural variation was conducted with gel electrophoresis for the presence of a specific open reading frame 7a(ORF7a) gene deletion within the "BE.9" lineages tracked. The analytical innovation of our method is the optimization of a simple intercalating dye-based qPCR assay through repositioning primers from the ARTIC v4.1 amplicon panel to detect large molecular patterns. This assay distinguishes between "BE.9" and "non-BE.9" lineages, particularly BQ.1, without the need for expensive probes or sequencing. The protocol was validated against lineage predictions from next-generation sequencing (NGS) using 525 paired samples, achieving 93.3% sensitivity, 95.1% specificity, and 92.4% agreement, as measured by Cohen's Kappa coefficient. Machine learning (ML) models were trained using the melting curves from intercalating dye-based qPCR of 1724 samples, enabling highly accurate lineage assignment. Among them, the support vector machine (SVM) model had the best performance and after fine-tuning showed ∼96.52% (333/345) accuracy in comparison to the test data set. Our integrated approach provides an adapted analytical method that is both cost-effective and scalable, suitable for rapid assessment of emerging variants, especially in resource-limited settings. In this work, the protocol is applied to improve the monitoring of SARS-CoV-2 sublineages but can be extended to track any key molecular signature, including large insertions and deletions (indels) commonly observed in pathogenic agent subtypes. By offering a complement to traditional sequencing methods and utilizing easily trainable machine learning algorithms, our methodology contributes to enhanced molecular surveillance strategies and supports global efforts in pandemic control.

A machine learning algorithm for the automatic classification of Phytophthora infestans genotypes into clonal lineages

Premise

The prompt categorization of Phytophthora infestans isolates into described clonal lineages is a key tool for the management of its associated disease, potato late blight. New isolates of this pathogen are currently classified by comparing their microsatellite genotypes with characterized clonal lineages, but an automated classification tool would greatly improve this process. Here, we developed a flexible machine learning-based classifier for P. infestans genotypes.

Methods

The performance of different machine learning algorithms in classifying P. infestans genotypes into its clonal lineages was preliminarily evaluated with decreasing amounts of training data. The four best algorithms were then evaluated using all collected genotypes.

Results

mlpML, cforest, nnet, and AdaBag performed best in the preliminary test, correctly classifying almost 100% of the genotypes. AdaBag performed significantly better than the others when tested using the complete data set (Tukey HSD P < 0.001). This algorithm was then implemented in a web application for the automated classification of P. infestans genotypes, which is freely available at https://github.com/cpatarroyo/genotypeclas.

Discussion

We developed a gradient boosting-based tool to automatically classify P. infestans genotypes into its clonal lineages. This could become a valuable resource for the prompt identification of clonal lineages spreading into new regions.

DNA melting analysis

Melting is a fundamental property of DNA that can be monitored by absorbance or fluorescence. PCR conveniently produces enough DNA to be directly monitored on real-time instruments with fluorescently labeled probes or dyes. Dyes monitor the entire PCR product, while probes focus on a specific locus within the amplicon. Advances in amplicon melting include high resolution instruments, saturating DNA dyes that better reveal multiple products, prediction programs for domain melting, barcode taxonomic identification, high speed microfluidic melting, and highly parallel digital melting. Most single base variants and small insertions or deletions can be genotyped by high resolution amplicon melting. High resolution melting also enables heterozygote scanning for any variant within a PCR product. A web application (uMelt, http://www.dna-utah.org) predicts amplicon melting curves with multiple domains, a useful tool for verifying intended products. Additional applications include methylation assessment, copy number determination and verification of sequence identity. When amplicon melting does not provide sufficient detail, unlabeled probes or snapback primers can be used instead of covalently labeled probes. DNA melting is a simple, inexpensive, and powerful tool with many research applications that is beginning to make its mark in clinical diagnostics.

Machine learning based DNA melt curve profiling enables automated novel genotype detection

Surveillance for genetic variation of microbial pathogens, both within and among species, plays an important role in informing research, diagnostic, prevention, and treatment activities for disease control. However, large-scale systematic screening for novel genotypes remains challenging in part due to technological limitations. Towards addressing this challenge, we present an advancement in universal microbial high resolution melting (HRM) analysis that is capable of accomplishing both known genotype identification and novel genotype detection. Specifically, this novel surveillance functionality is achieved through time-series modeling of sequence-defined HRM curves, which is uniquely enabled by the large-scale melt curve datasets generated using our high-throughput digital HRM platform. Taking the detection of bacterial genotypes as a model application, we demonstrate that our algorithms accomplish an overall classification accuracy over 99.7% and perform novelty detection with a sensitivity of 0.96, specificity of 0.96 and Youden index of 0.92. Since HRM-based DNA profiling is an inexpensive and rapid technique, our results add support for the feasibility of its use in surveillance applications.

Improving bacteria identification from digital melt assay via oligonucleotide-based temperature calibration

BACKGROUND

Bacterial infections, especially polymicrobial infections, remain a threat to global health and require advances in diagnostic technologies for timely and accurate identification of all causative species. Digital melt - microfluidic chip-based digital PCR combined with high resolution melt (HRM) - is an emerging method for identification and quantification of polymicrobial bacterial infections. Despite advances in recent years, existing digital melt instrumentation often delivers nonuniform temperatures across digital chips, resulting in nonuniform digital melt curves for individual bacterial species. This nonuniformity can lead to inaccurate species identification and reduce the capacity for differentiating bacterial species with similar digital melt curves.

RESULTS

We introduce herein a new temperature calibration method for digital melt by incorporating an unamplified, synthetic DNA fragment with a known melting temperature as a calibrator. When added at a tuned concentration to an established digital melt assay amplifying the commonly targeted 16S V1 - V6 region, this calibrator produced visible low temperature calibrator melt curves across-chip along with the target bacterial melt curves. This enables alignment of the bacterial melt curves and correction of heating-induced nonuniformities. Using this calibration method, we were able to improve the uniformity of digital melt curves from three causative species of bacteria. Additionally, we assessed calibration's effects on identification accuracy by performing machine learning identification of three polymicrobial mixtures comprised of two bacteria with similar digital melt curves in different ratios. Calibration greatly improved mixture composition prediction.

SIGNIFICANCE

To the best of our knowledge, this work represents the first DNA calibrator-supplemented assay and calibration method for nanoarray digital melt. Our results suggest that this calibration method can be flexibly used to improve identification accuracy and reduce melt curve variabilities across a variety of pathogens and assays. Therefore, this calibration method has the potential to elevate the diagnostic capabilities of digital melt toward polymicrobial bacterial infections and other infectious diseases.

Harnessing Variabilities in Digital Melt Curves for Accurate Identification of Bacteria

Digital PCR combined with high resolution melt (HRM) is an emerging method for identifying pathogenic bacteria with single cell resolution via species-specific digital melt curves. Currently, the development of such digital PCR-HRM assays entails first identifying PCR primers to target hypervariable gene regions within the target bacteria panel, next performing bulk-based PCR-HRM to examine whether the resulting species-specific melt curves possess sufficient interspecies variability (i.e., variability between bacterial species), and then digitizing the bulk-based PCR-HRM assays with melt curves that have high interspecies variability via microfluidics. In this work, we first report our discovery that the current development workflow can be inadequate because a bulk-based PCR-HRM assay that produces melt curves with high interspecies variability can, in fact, lead to a digital PCR-HRM assay that produces digital melt curves with unwanted intraspecies variability (i.e., variability within the same bacterial species), consequently hampering bacteria identification accuracy. Our subsequent investigation reveals that such intraspecies variability in digital melt curves can arise from PCR primers that target nonidentical gene copies or amplify nonspecifically. We then show that computational in silico HRM opens a window to inspect both interspecies and intraspecies variabilities and thus provides the missing link between bulk-based PCR-HRM and digital PCR-HRM. Through this new development workflow, we report a new digital PCR-HRM assay with improved bacteria identification accuracy. More broadly, this work can serve as the foundation for enhancing the development of future digital PCR-HRM assays toward identifying causative pathogens and combating infectious diseases.

An update on current and novel molecular diagnostics for the diagnosis of invasive fungal infections

BACKGROUND

Invasive fungal infections cause millions of infections annually, but diagnosis remains challenging. There is an increased need for low-cost, easy to use, highly sensitive and specific molecular assays that can differentiate between colonized and pathogenic organisms from different clinical specimens.

AREAS COVERED

We reviewed the literature evaluating the current state of molecular diagnostics for invasive fungal infections, focusing on current and novel molecular tests such as polymerase chain reaction (PCR), digital PCR, high-resolution melt (HRM), and metagenomics/next generation sequencing (mNGS).

EXPERT OPINION

PCR is highly sensitive and specific, although performance can be impacted by prior/concurrent antifungal use. PCR assays can identify mutations associated with antifungal resistance, non-Aspergillus mold infections, and infections from endemic fungi. HRM is a rapid and highly sensitive diagnostic modality that can identify a wide range of fungal pathogens, including down to the species level, but multiplex assays are limited and HRM is currently unavailable in most healthcare settings, although universal HRM is working to overcome this limitation. mNGS offers a promising approach for rapid and hypothesis-free diagnosis of a wide range of fungal pathogens, although some drawbacks include limited access, variable performance across platforms, the expertise and costs associated with this method, and long turnaround times in real-world settings.

Adaptive Filtering Framework to Remove Nonspecific and Low-Efficiency Reactions in Multiplex Digital PCR Based on Sigmoidal Trends

Real-time digital polymerase chain reaction (qdPCR) coupled with machine learning (ML) methods has shown the potential to unlock scientific breakthroughs, particularly in the field of molecular diagnostics for infectious diseases. One promising application of this emerging field explores single fluorescent channel PCR multiplex by extracting target-specific kinetic and thermodynamic information contained in amplification curves, also known as data-driven multiplexing. However, accurate target classification is compromised by the presence of undesired amplification events and not ideal reaction conditions. Therefore, here, we proposed a novel framework to identify and filter out nonspecific and low-efficient reactions from qdPCR data using outlier detection algorithms purely based on sigmoidal trends of amplification curves. As a proof-of-concept, this framework is implemented to improve the classification performance of the recently reported data-driven multiplexing method called amplification curve analysis (ACA), using available published data where the ACA is demonstrated to screen carbapenemase-producing organisms in clinical isolates. Furthermore, we developed a novel strategy, named adaptive mapping filter (AMF), to adjust the percentage of outliers removed according to the number of positive counts in qdPCR. From an overall total of 152,000 amplification events, 116,222 positive amplification reactions were evaluated before and after filtering by comparing against melting peak distribution, proving that abnormal amplification curves (outliers) are linked to shifted melting distribution or decreased PCR efficiency. The ACA was applied to assess classification performance before and after AMF, showing an improved sensitivity of 1.2% when using inliers compared to a decrement of 19.6% when using outliers (p-value < 0.0001), removing 53.5% of all wrong melting curves based only on the amplification shape. This work explores the correlation between the kinetics of amplification curves and the thermodynamics of melting curves, and it demonstrates that filtering out nonspecific or low-efficient reactions can significantly improve the classification accuracy for cutting-edge multiplexing methodologies.

Diagnosis of Bloodstream Infections: An Evolution of Technologies towards Accurate and Rapid Identification and Antibiotic Susceptibility Testing

Bloodstream infections (BSI) are a leading cause of death worldwide. The lack of timely and reliable diagnostic practices is an ongoing issue for managing BSI. The current gold standard blood culture practice for pathogen identification and antibiotic susceptibility testing is time-consuming. Delayed diagnosis warrants the use of empirical antibiotics, which could lead to poor patient outcomes, and risks the development of antibiotic resistance. Hence, novel techniques that could offer accurate and timely diagnosis and susceptibility testing are urgently needed. This review focuses on BSI and highlights both the progress and shortcomings of its current diagnosis. We surveyed clinical workflows that employ recently approved technologies and showed that, while offering improved sensitivity and selectivity, these techniques are still unable to deliver a timely result. We then discuss a number of emerging technologies that have the potential to shorten the overall turnaround time of BSI diagnosis through direct testing from whole blood-while maintaining, if not improving-the current assay's sensitivity and pathogen coverage. We concluded by providing our assessment of potential future directions for accelerating BSI pathogen identification and the antibiotic susceptibility test. While engineering solutions have enabled faster assay turnaround, further progress is still needed to supplant blood culture practice and guide appropriate antibiotic administration for BSI patients.

Coupling Machine Learning and High Throughput Multiplex Digital PCR Enables Accurate Detection of Carbapenem-Resistant Genes in Clinical Isolates

Rapid and accurate identification of patients colonised with carbapenemase-producing organisms (CPOs) is essential to adopt prompt prevention measures to reduce the risk of transmission. Recent studies have demonstrated the ability to combine machine learning (ML) algorithms with real-time digital PCR (dPCR) instruments to increase classification accuracy of multiplex PCR assays when using synthetic DNA templates. We sought to determine if this novel methodology could be applied to improve identification of the five major carbapenem-resistant genes in clinical CPO-isolates, which would represent a leap forward in the use of PCR-based data-driven diagnostics for clinical applications. We collected 253 clinical isolates (including 221 CPO-positive samples) and developed a novel 5-plex PCR assay for detection of bla_IMP, bla_KPC, bla_NDM, bla_OXA-48, and bla_VIM. Combining the recently reported ML method “Amplification and Melting Curve Analysis” (AMCA) with the abovementioned multiplex assay, we assessed the performance of the AMCA methodology in detecting these genes. The improved classification accuracy of AMCA relies on the usage of real-time data from a single-fluorescent channel and benefits from the kinetic/thermodynamic information encoded in the thousands of amplification events produced by high throughput real-time dPCR. The 5-plex showed a lower limit of detection of 10 DNA copies per reaction for each primer set and no cross-reactivity with other carbapenemase genes. The AMCA classifier demonstrated excellent predictive performance with 99.6% (CI 97.8–99.9%) accuracy (only one misclassified sample out of the 253, with a total of 160,041 positive amplification events), which represents a 7.9% increase (p-value <0.05) compared to conventional melting curve analysis. This work demonstrates the use of the AMCA method to increase the throughput and performance of state-of-the-art molecular diagnostic platforms, without hardware modifications and additional costs, thus potentially providing substantial clinical utility on screening patients for CPO carriage.

High resolution melting assay as a reliable method for diagnosing drug-resistant TB cases: a systematic review and meta-analysis

BACKGROUND

Tuberculosis (TB) is one of the most contagious infectious diseases worldwide. Currently, drug-resistant Mycobacterium tuberculosis (Mtb) isolates are considered as one of the main challenges in the global TB control strategy. Rapid detection of resistant strains effectively reduces morbidity and mortality of world's population. Although both culture and conventional antibiotic susceptibility testing are time-consuming, recent studies have shown that high resolution melting (HRM) assay can be used to determine the types of antibiotic resistance. In the present meta-analysis, we evaluated the discriminative power of HRM in detecting all drug-resistance cases of TB.

METHODS

A systematic search was performed using databases such as Cochrane Library, Scopus, PubMed, Web of Science, and Google Scholar. Related studies on the effect of HRM in the diagnosis of drug-resistant (DR) TB cases were retrieved by April 2021. We used Meta-Disc software to evaluate the pooled diagnostic sensitivity and specificity of HRM for the detection of each type of drug-resistant cases. Finally, diagnostic value of HRM was characterized by summary receiver operating characteristic (SROC) curve and the area under the curve (AUC) method.

RESULTS

Overall 47 studies (4,732 Mtb isolates) met our criteria and were included in the present meta-analysis. Sensitivity, specificity, and AUC of HRM were measured for antibiotics such as isoniazid (93%, 98%, 0.987), rifampin (94%, 97%, 0963), ethambutol (82%, 87%, 0.728), streptomycin (82%, 95%, 0.957), pyrazinamide (72%, 84%, 0.845), fluoroquinolones (86%, 99%, 0.997), MDR-TB (90%, 98%, 0.989), and pan-drug-resistant TB (89%, 95%, 0.973).

CONCLUSIONS

The HRM assay has high accuracy for the identification of drug-resistant TB, particularly firs-line anti-TB drugs. Therefore, this method is considered as an alternative option for the rapid diagnosis of DR-TB cases. However, due to heterogeneity of included studies, the results of HRM assays should be interpreted based on conventional drug susceptibility testing.

Convolutional neural network analysis of recurrence plots for high resolution melting classification

BACKGROUND AND OBJECTIVE

High resolution melting (HRM) analysis is a rapid and correct method for identification of species, such as, microorganism, bacteria, yeast, virus, etc. HRM data are produced using real-time polymerase chain reaction (PCR) and unique for each species. Analysis of the HRM data is important for several applications, such as, for detection of diseases (e.g., influenza, zika virus, SARS-Cov-2 and Covid-19 diseases) in health, for identification of spoiled foods in food industry, for analysis of crime scene evidence in forensic investigation, etc. However, the characteristics of the HRM data can change due to the experimental conditions or instrumental settings. In addition, it becomes laborious and time-consuming process as the number of samples increases. Because of these reasons, the analysis and classification of the HRM data become challenging for species which have similar characteristics.

METHODS

To improve the classification accuracy of HRM data, we propose to use image (visual) representation of HRM data, which we call HRM images, that are generated using recurrence plots, and propose convolutional neural network (CNN) based models for classifying HRM images. In this study, two different types of recurrence plots are generated, which are black-white recurrence plots (BW-RP) and gray scale recurrence plots (GS-RP) and four different CNN models are proposed for classifying HRM data.

RESULTS

The classification performance of the proposed methods are evaluated based on average classification accuracy and F1 score, specificity, recall, and precision values for each yeast species. When BW-RP representation of HRM data is used as input to the CNN models, the best classification accuracy of 95.2% is obtained. The classification accuracies of CNN models for melting curve and GS-RP data representations of HRM data are 90.13% and 86.13%, respectively. The classification accuracy of support vector machines (SVM) model that take melting curve representation of HRM data is 86.53%. Moreover, when BW-RP representation of HRM data is used as input to the CNN models, the F1 score, specificity, recall and precision values are the highest for almost all of species.

CONCLUSIONS

Experimental results show that using BW-RP representation of HRM data improved the classification accuracy of HRM data and CNN models that take these images as input outperformed CNN models that take melting curve and GS-RP representations of HRM data as inputs and SVM model that take melting curve representation of HRM data as input.

STRide probes: Single-labeled short tandem repeat identification probes

The demand for forensic DNA profiling at the crime scene or at police stations is increasing. DNA profiling is currently performed in specialized laboratories by PCR amplification of Short Tandem Repeats (STR) followed by amplicon sizing using capillary electrophoresis. The need for bulky equipment to identify alleles after PCR presents a challenge for shifting to a decentralized workflow. We devised a novel hybridization-based STR-genotyping method, using Short Tandem Repeat Identification (STRide) probes, which could help tackle this issue. STRide probes are fluorescently labeled oligonucleotides that rely on the quenching properties of guanine on fluorescein derivatives. Mismatches between STRide probes and amplicons can be detected by melting curve analysis after asymmetric PCR. The functionality of the STRide probes was demonstrated by analyzing synthetic DNA samples for the D16S539 locus. Next, STRide probes were developed for five different CODIS core loci (D16S539, TH01, TPOX, FGA, and D7S820). These probes were validated by analyzing 13 human DNA samples. Successful genotyping was obtained using inputs as low as 31 pg of DNA, demonstrating high sensitivity. The STRide probes are ideally suited to be implemented in a microarray and present an important step towards a portable device for fast on-site forensic DNA fingerprinting.

Data-driven noise modeling of digital DNA melting analysis enables prediction of sequence discriminating power

MOTIVATION

The need to rapidly screen complex samples for a wide range of nucleic acid targets, like infectious diseases, remains unmet. Digital High-Resolution Melt (dHRM) is an emerging technology with potential to meet this need by accomplishing broad-based, rapid nucleic acid sequence identification. Here, we set out to develop a computational framework for estimating the resolving power of dHRM technology for defined sequence profiling tasks. By deriving noise models from experimentally generated dHRM datasets and applying these to in silico predicted melt curves, we enable the production of synthetic dHRM datasets that faithfully recapitulate real-world variations arising from sample and machine variables. We then use these datasets to identify the most challenging melt curve classification tasks likely to arise for a given application and test the performance of benchmark classifiers.

RESULTS

This toolbox enables the in silico design and testing of broad-based dHRM screening assays and the selection of optimal classifiers. For an example application of screening common human bacterial pathogens, we show that human pathogens having the most similar sequences and melt curves are still reliably identifiable in the presence of experimental noise. Further, we find that ensemble methods outperform whole series classifiers for this task and are in some cases able to resolve melt curves with single-nucleotide resolution.

AVAILABILITY

Data and code available on https://github.com/lenlan/dHRM-noise-modeling.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

High-Level Multiplexing in Digital PCR with Intercalating Dyes by Coupling Real-Time Kinetics and Melting Curve Analysis

Digital polymerase chain reaction (dPCR) is a mature technique that has enabled scientific breakthroughs in several fields. However, this technology is primarily used in research environments with high-level multiplexing, representing a major challenge. Here, we propose a novel method for multiplexing, referred to as amplification and melting curve analysis (AMCA), which leverages the kinetic information in real-time amplification data and the thermodynamic melting profile using an affordable intercalating dye (EvaGreen). The method trains a system composed of supervised machine learning models for accurate classification, by virtue of the large volume of data from dPCR platforms. As a case study, we develop a new 9-plex assay to detect mobilized colistin resistant genes as clinically relevant targets for antimicrobial resistance. Over 100,000 amplification events have been analyzed, and for the positive reactions, the AMCA approach reports a classification accuracy of 99.33 ± 0.13%, an increase of 10.0% over using melting curve analysis. This work provides an affordable method of high-level multiplexing without fluorescent probes, extending the benefits of dPCR in research and clinical settings.

Application of machine learning algorithm and modified high resolution DNA melting curve analysis for molecular subtyping of Salmonella isolates from various epidemiological backgrounds in northern Thailand

Food poisoning from consumption of food contaminated with non-typhoidal Salmonella spp. is a global problem. A modified high resolution DNA melting curve analysis (m-HRMa) was introduced to provide effective discrimination among closely related HRM curves of amplicons generated from selected Salmonella genome sequences enabled Salmonella spp. to be classified into discrete clusters. Combination of m-HRMa with serogroup identification (ms-HRMa) helped improve assignment of Salmonella spp. into clusters. In addition, a machine learning (dynamic time warping) algorithm (DTW) was employed to provide a simple and rapid protocol for clustering analysis as well as to create phylogeny tree of Salmonella strains (n = 40) collected from home, farms and slaughter houses in northern Thailand. Applications of DTW and ms-HRMa clustering analyses were capable of generating molecular signatures of the Salmonella isolates, resulting in 25 ms-HRM and 28 DTW clusters compared to 14 clusters from a standard HRM analysis, and the combination of both analyses permitted molecular subtyping of each Salmonella isolate. Results from DTW and ms-HRMa cluster analyses were in good agreement with that obtained from enterobacterial repetitive intergenic consensus sequence PCR clustering. While conventional serotyping of Clusters 1 and 2 revealed six different Salmonella serotypes, the majority being S. Weltevraden, the new Salmonella subtyping protocol identified five S. Weltevraden subtypes with S.Weltevreden subtype DTW4-M1 being predominant. Based on knowledge of the sources of Salmonella subtypes, transmission of S. Weltevraden in northern Thailand was likely to be farm-to-farm through contaminated chicken stool. In conclusion, the rapid, robust and specific Salmonella subtyping developed in the study can be performed in a local setting, enabling swift control and preventive measures to be initiated against potential epidemics of salmonellosis.

Improving Quantitative Power in Digital PCR through Digital High-Resolution Melting

Applying digital PCR (dPCR) technology to challenging clinical and industrial detection tasks has become more prevalent because of its capability for absolute quantification and rare target detection. However, practices learned from quantitative PCR (qPCR) that promote assay robustness and wide-ranging utility are not readily applied in dPCR. These include internal amplification controls to account for false-negative reactions and amplicon high-resolution melt (HRM) analysis to distinguish true positives from false positives. Incorporation of internal amplification controls in dPCR is challenging because of the limited fluorescence channels available on most machines, and the application of HRM analysis is hindered by the separation of heating and imaging functions on most dPCR systems. We use a custom digital HRM platform to assess the utility of HRM-based approaches for mitigation of false positives and false negatives in dPCR. We show that detection of an exogenous internal control using dHRM analysis reduces the inclusion of false-negative partitions, changing the calculated DNA concentration up to 52%. The integration of dHRM analysis enables classification of partitions that would otherwise be considered ambiguous "rain," which accounts for up to ∼3% and ∼10% of partitions in intercalating dye and hydrolysis probe dPCR, respectively. We focused on developing an internal control method that would be compatible with broad-based microbial detection in dPCR-dHRM. Our approach can be applied to a number of DNA detection methods including microbial profiling and may advance the utility of dPCR in clinical applications where accurate quantification is imperative.

Three-gene PCR and high-resolution melting analysis for differentiating vertebrate species mitochondrial DNA for biodiversity research and complementing forensic surveillance

Reliable molecular identification of vertebrate species from morphologically unidentifiable tissue is critical for the prosecution of illegally-traded wildlife products, conservation-based biodiversity research, and identification of blood-meal hosts of hematophagous invertebrates. However, forensic identification of vertebrate tissue relies on sequencing of the mitochondrial cytochrome oxidase I (COI) 'barcode' gene, which remains costly for purposes of screening large numbers of unknown samples during routine surveillance. Here, we adapted a rapid, low-cost approach to differentiate 10 domestic and 24 wildlife species that are common in the East African illegal wildlife products trade based on their unique high-resolution melting profiles from COI, cytochrome b, and 16S ribosomal RNA gene PCR products. Using the approach, we identified (i) giraffe among covertly sampled meat from Kenyan butcheries, and (ii) forest elephant mitochondrial sequences among savannah elephant reference samples. This approach is being adopted for high-throughput pre-screening of potential bushmeat samples in East African forensic science pipelines.

Nanoarray Digital Polymerase Chain Reaction with High-Resolution Melt for Enabling Broad Bacteria Identification and Pheno-Molecular Antimicrobial Susceptibility Test

Toward combating infectious diseases caused by pathogenic bacteria, there remains an unmet need for diagnostic tools that can broadly identify the causative bacteria and determine their antimicrobial susceptibilities from complex and even polymicrobial samples in a timely manner. To address this need, a microfluidic and machine-learning-based platform that performs broad bacteria identification (ID) and rapid yet reliable antimicrobial susceptibility testing (AST) is developed. Specifically, this platform builds on "pheno-molecular AST", a strategy that transforms nucleic acid amplification tests (NAATs) into phenotypic AST through quantitative detection of bacterial genomic replication, and utilizes digital polymerase chain reaction (PCR) and digital high-resolution melt (HRM) to quantify and identify bacterial DNA molecules. Bacterial species are identified using integrated experiment-machine learning algorithm via HRM profiles. Digital DNA quantification allows for rapid growth measurement that reflects susceptibility profiles of each bacterial species within only 30 min of antibiotic exposure. As a demonstration, multiple bacterial species and their susceptibility profiles in a spiked-in polymicrobial urine specimen were correctly identified with a total turnaround time of ∼4 h. With further development and clinical validation, this platform holds the potential for improving clinical diagnostics and enabling targeted antibiotic treatments.

Emerging Analytical Techniques for Rapid Pathogen Identification and Susceptibility Testing

In the face of looming threats from multi-drug resistant microorganisms, there is a growing need for technologies that will enable rapid identification and drug susceptibility profiling of these pathogens in health care settings. In particular, recent progress in microfluidics and nucleic acid amplification is pushing the boundaries of timescale for diagnosing bacterial infections. With a diverse range of techniques and parallel developments in the field of analytical chemistry, an integrative perspective is needed to understand the significance of these developments. This review examines the scope of new developments in assay technologies grouped by key enabling domains of research. First, we examine recent development in nucleic acid amplification assays for rapid identification and drug susceptibility testing in bacterial infections. Next, we examine advances in microfluidics that facilitate acceleration of diagnostic assays via integration and scale. Lastly, recentdevelopments in biosensor technologies are reviewed. We conclude this review with perspectives on the use of emerging concepts to develop paradigm-changing assays.

A 'culture' shift: Application of molecular techniques for diagnosing polymicrobial infections

With the advancement of microbiological discovery, it is evident that many infections, particularly bloodstream infections, are polymicrobial in nature. Consequently, new challenges have emerged in identifying the numerous etiologic organisms in an accurate and timely manner using the current diagnostic standard. Various molecular diagnostic methods have been utilized as an effort to provide a fast and reliable identification in lieu or parallel to the conventional culture-based methods. These technologies are mostly based on nucleic acid, proteins, or physical properties of the pathogens with differing advantages and limitations. This review evaluates the different molecular methods and technologies currently available to diagnose polymicrobial infections, which will help determine the most appropriate option for future diagnosis.

Digital High-Resolution Melt Platform for Rapid and Parallelized Molecule-by-Molecule Genetic Profiling

This work presents a microfluidic Digital High-Resolution Melt platform for absolute quantitation and sensitive detection of locus-specific sequence variations on a molecule-by-molecule basis. The platform provides a facile means for assessment of hundreds to thousands of single DNA copies by digitizing template molecules in a 4096 1-nL array microfluidic device and observing the sequence-dependent fluorescence changes during temperature ramping. The analytical capability of this platform is demonstrated in several applications, such as digital assay characterization, detection and assessment of DNA methylation heterogeneity, and detection of rare biomarkers at frequencies as low as 0.0005% target to background molecules.

Integrated Bacterial Identification and Antimicrobial Susceptibility Testing for Polymicrobial Infections Using Digital PCR and Digital High-Resolution Melt in a Microfluidic Array Platform

In diagnosing bacterial infection, rapid bacterial identification (ID) and antimicrobial susceptibility testing (AST) are critical to clinicians in order to provide an effective treatment in a timely manner. The gold standard, culture-based approach provides both ID and antimicrobial susceptibility but requires several days of turnaround time. Especially in polymicrobial infections, where there are more than one organisms interacting collectively that can complicate the treatment. Here, we demonstrate a rapid bacterial diagnostic approach that is capable of bacterial ID/AST in heterogeneous samples within less than 4 hours by using digital PCR (dPCR) and digital high-resolution melt via microfluidic devices. By utilizing dPCR, we are able to quantify amount of nucleic acid, which correlates to phenotypic responses of {\bf individual pathogens in a mixed sample and also shorten the required time of antibiotic exposure. In addition, we employ a machine learning algorithm to automatically identify bacterial species based on melt profiles of 16S rRNA gene.

Recognition Tunneling of Canonical and Modified RNA Nucleotides for Their Identification with the Aid of Machine Learning

In the present study, we demonstrate a tunneling nanogap technique to identify individual RNA nucleotides, which can be used as a mechanism to read the nucleobases for direct sequencing of RNA in a solid-state nanopore. The tunneling nanogap is composed of two electrodes separated by a distance of <3 nm and functionalized with a recognition molecule. When a chemical entity is captured in the gap, it generates electron tunneling currents, a process we call recognition tunneling (RT). Using RT nanogaps created in a scanning tunneling microscope (STM), we acquired the electron tunneling signals for the canonical and two modified RNA nucleotides. To call the individual RNA nucleotides from the RT data, we adopted a machine learning algorithm, support vector machine (SVM), for the data analysis. Through the SVM, we were able to identify the individual RNA nucleotides and distinguish them from their DNA counterparts with reasonably high accuracy. Since each RNA nucleoside contains a hydroxyl group at the 2'-position of its sugar ring in an RNA strand, it allows for the formation of a tunneling junction at a larger nanogap compared to the DNA nucleoside in a DNA strand, which lacks the 2' hydroxyl group. It also proves advantageous for the manufacture of RT devices. This study is a proof-of-principle demonstration for the development of an RT nanopore device for directly sequencing single RNA molecules, including those bearing modifications.

A High-Resolution Digital DNA Melting Platform for Robust Sequence Profiling and Enhanced Genotype Discrimination

DNA melting analysis provides a rapid method for genotyping a target amplicon directly after PCR amplification. To transform melt genotyping into a broad-based profiling approach for heterogeneous samples, we previously proposed the integration of universal PCR and melt analysis with digital PCR. Here, we advanced this concept by developing a high-resolution digital melt platform with precise thermal control to accomplish reliable, high-throughput heat ramping of microfluidic chip digital PCR reactions. Using synthetic DNA oligos with defined melting temperatures, we characterized sources of melting variability and minimized run-to-run variations. Within-run comparisons throughout a 20,000-reaction chip revealed that high-melting-temperature sequences were significantly less prone to melt variation. Further optimization using bacterial 16S amplicons revealed a strong dependence of the number of melting transitions on the heating rate during curve generation. These studies show that reliable high-resolution melt curve genotyping can be achieved in digital, picoliter-scale reactions and demonstrate that rate-dependent melt signatures may be useful for enhancing automated melt genotyping.

High-Resolution Melting Analysis for Rapid Detection of Sequence Type 131 Escherichia coli

Escherichia coli isolates belonging to the sequence type 131 (ST131) clonal complex have been associated with the global distribution of fluoroquinolone and β-lactam resistance. Whole-genome sequencing and multilocus sequence typing identify sequence type but are expensive when evaluating large numbers of samples. This study was designed to develop a cost-effective screening tool using high-resolution melting (HRM) analysis to differentiate ST131 from non-ST131 E. coli in large sample populations in the absence of sequence analysis. The method was optimized using DNA from 12 E. coli isolates. Singleplex PCR was performed using 10 ng of DNA, Type-it HRM buffer, and multilocus sequence typing primers and was followed by multiplex PCR. The amplicon sizes ranged from 630 to 737 bp. Melt temperature peaks were determined by performing HRM analysis at 0.1°C resolution from 50 to 95°C on a Rotor-Gene Q 5-plex HRM system. Derivative melt curves were compared between sequence types and analyzed by principal component analysis. A blinded study of 191 E. coli isolates of ST131 and unknown sequence types validated this methodology. This methodology returned 99.2% specificity (124 true negatives and 1 false positive) and 100% sensitivity (66 true positives and 0 false negatives). This HRM methodology distinguishes ST131 from non-ST131 E. coli without sequence analysis. The analysis can be accomplished in about 3 h in any laboratory with an HRM-capable instrument and principal component analysis software. Therefore, this assay is a fast and cost-effective alternative to sequencing-based ST131 identification.

Massively parallel digital high resolution melt for rapid and absolutely quantitative sequence profiling

In clinical diagnostics and pathogen detection, profiling of complex samples for low-level genotypes represents a significant challenge. Advances in speed, sensitivity, and extent of multiplexing of molecular pathogen detection assays are needed to improve patient care. We report the development of an integrated platform enabling the identification of bacterial pathogen DNA sequences in complex samples in less than four hours. The system incorporates a microfluidic chip and instrumentation to accomplish universal PCR amplification, High Resolution Melting (HRM), and machine learning within 20,000 picoliter scale reactions, simultaneously. Clinically relevant concentrations of bacterial DNA molecules are separated by digitization across 20,000 reactions and amplified with universal primers targeting the bacterial 16S gene. Amplification is followed by HRM sequence fingerprinting in all reactions, simultaneously. The resulting bacteria-specific melt curves are identified by Support Vector Machine learning, and individual pathogen loads are quantified. The platform reduces reaction volumes by 99.995% and achieves a greater than 200-fold increase in dynamic range of detection compared to traditional PCR HRM approaches. Type I and II error rates are reduced by 99% and 100% respectively, compared to intercalating dye-based digital PCR (dPCR) methods. This technology could impact a number of quantitative profiling applications, especially infectious disease diagnostics.

Microbial Typing by Machine Learned DNA Melt Signatures

There is still an ongoing demand for a simple broad-spectrum molecular diagnostic assay for pathogenic bacteria. For this purpose, we developed a single-plex High Resolution Melt (HRM) assay that generates complex melt curves for bacterial identification. Using internal transcribed spacer (ITS) region as the phylogenetic marker for HRM, we observed complex melt curve signatures as compared to 16S rDNA amplicons with enhanced interspecies discrimination. We also developed a novel Naïve Bayes curve classification algorithm with statistical interpretation and achieved 95% accuracy in differentiating 89 bacterial species in our library using leave-one-out cross-validation. Pilot clinical validation of our method correctly identified the etiologic organisms at the species-level in 59 culture-positive mono-bacterial blood culture samples with 90% accuracy. Our findings suggest that broad bacterial sequences may be simply, reliably and automatically profiled by ITS HRM assay for clinical adoption.

Arbovirus and insect-specific virus discovery in Kenya by novel six genera multiplex high-resolution melting analysis

A broad diversity of arthropod-borne viruses (arboviruses) of global health concern are endemic to East Africa, yet most surveillance efforts are limited to just a few key viral pathogens. Additionally, estimates of arbovirus diversity in the tropics are likely to be underestimated as their discovery has lagged significantly over past decades due to limitations in fast and sensitive arbovirus identification methods. Here, we developed a nearly pan-arbovirus detection assay that uses high-resolution melting (HRM) analysis of RT-PCR products from highly multiplexed assays to differentiate broad diversities of arboviruses. We differentiated 15 viral culture controls and seven additional synthetic viral DNA sequence controls, within Flavivirus, Alphavirus, Nairovirus, Phlebovirus, Orthobunyavirus and Thogotovirus genera. Among Bunyamwera, sindbis, dengue and Thogoto virus serial dilutions, detection by multiplex RT-PCR-HRM was comparable to the gold standard Vero cell plaque assays. We applied our low-cost method for enhanced broad-range pathogen surveillance from mosquito samples collected in Kenya and identified diverse insect-specific viruses, including a new clade in anopheline mosquitoes, and Wesselsbron virus, an arbovirus that can cause viral haemorrhagic fever in humans and has not previously been isolated in Kenya, in Culex spp. and Anopheles coustani mosquitoes. Our findings demonstrate how multiplex RT-PCR-HRM can identify novel viral diversities and potential disease threats that may not be included in pathogen detection panels of routine surveillance efforts. This approach can be adapted to other pathogens to enhance disease surveillance and pathogen discovery efforts, as well as the study of pathogen diversity and viral evolutionary ecology.

Nested Machine Learning Facilitates Increased Sequence Content for Large-Scale Automated High Resolution Melt Genotyping

High Resolution Melt (HRM) is a versatile and rapid post-PCR DNA analysis technique primarily used to differentiate sequence variants among only a few short amplicons. We recently developed a one-vs-one support vector machine algorithm (OVO SVM) that enables the use of HRM for identifying numerous short amplicon sequences automatically and reliably. Herein, we set out to maximize the discriminating power of HRM + SVM for a single genetic locus by testing longer amplicons harboring significantly more sequence information. Using universal primers that amplify the hypervariable bacterial 16 S rRNA gene as a model system, we found that long amplicons yield more complex HRM curve shapes. We developed a novel nested OVO SVM approach to take advantage of this feature and achieved 100% accuracy in the identification of 37 clinically relevant bacteria in Leave-One-Out-Cross-Validation. A subset of organisms were independently tested. Those from pure culture were identified with high accuracy, while those tested directly from clinical blood bottles displayed more technical variability and reduced accuracy. Our findings demonstrate that long sequences can be accurately and automatically profiled by HRM with a novel nested SVM approach and suggest that clinical sample testing is feasible with further optimization.

Automated Classification and Cluster Visualization of Genotypes Derived from High Resolution Melt Curves

Introduction

High Resolution Melting (HRM) following PCR has been used to identify DNA genotypes. Fluorescent dyes bounded to double strand DNA lose their fluorescence with increasing temperature, yielding different signatures for different genotypes. Recent software tools have been made available to aid in the distinction of different genotypes, but they are not fully automated, used only for research purposes, or require some level of interaction or confirmation from an analyst.

Materials and Methods

We describe a fully automated machine learning software algorithm that classifies unknown genotypes. Dynamic melt curves are transformed to multidimensional clusters of points whereby a training set is used to establish the distribution of genotype clusters. Subsequently, probabilistic and statistical methods were used to classify the genotypes of unknown DNA samples on 4 different assays (40 VKORC1, CYP2C9*2, CYP2C9*3 samples in triplicate, and 49 MTHFR c.665C>T samples in triplicate) run on the Roche LC480. Melt curves of each of the triplicates were genotyped separately.

Results

Automated genotyping called 100% of VKORC1, CYP2C9*3 and MTHFR c.665C>T samples correctly. 97.5% of CYP2C9*2 melt curves were genotyped correctly with the remaining 2.5% given a no call due to the inability to decipher 3 melt curves in close proximity as either homozygous mutant or wild-type with greater than 99.5% posterior probability.

Conclusions

We demonstrate the ability to fully automate DNA genotyping from HRM curves systematically and accurately without requiring any user interpretation or interaction with the data. Visualization of genotype clusters and quantification of the expected misclassification rate is also available to provide feedback to assay scientists and engineers as changes are made to the assay or instrument.