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.

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.

Sensitive detection and serovar differentiation of typhoidal and nontyphoidal Salmonella enterica species using 16S rRNA Gene PCR coupled with high-resolution melt analysis.

Salmonella enterica species infections are a significant public health problem causing high morbidity rates worldwide and high mortality rates in the developing world. These infections are not always rapidly diagnosed as a cause of bloodstream infections because of the limitations of blood culture, which greatly affects clinical care as a result of treatment delays. A molecular diagnostic assay that could rapidly detect and identify S. enterica species infections as a cause of sepsis is needed. Nine typhoidal and nontyphoidal S. enterica serovars were used to establish the limit of detection (LOD) of a previously published 16S rRNA gene PCR (16S PCR) in mock whole blood specimens. In addition, 16 typhoidal and nontyphoidal S. enterica serovars were used to evaluate the serovar differentiation capability of 16S PCR coupled with high-resolution melt analysis. The overall LOD of 16S PCR for the nine typhoidal and nontyphoidal S. enterica serovars analyzed was <10 colony-forming units per milliliter (CFU/mL) in mock whole blood specimens, with the lowest and highest LOD at <1 CFU/mL and 9 CFU/mL, respectively. By high-resolution melt analysis, the typhoidal and nontyphoidal S. enterica serovar groups analyzed each generated a unique grouping code, allowing for serovar-level identification. 16S PCR coupled with high-resolution melt analysis could be a useful molecular diagnostic that could enhance the current diagnostic, treatment, and surveillance methods of S. enterica bloodstream infections.

Universal digital high-resolution melt: a novel approach to broad-based profiling of heterogeneous biological samples.

Comprehensive profiling of nucleic acids in genetically heterogeneous samples is important for clinical and basic research applications. Universal digital high-resolution melt (U-dHRM) is a new approach to broad-based PCR diagnostics and profiling technologies that can overcome issues of poor sensitivity due to contaminating nucleic acids and poor specificity due to primer or probe hybridization inaccuracies for single nucleotide variations. The U-dHRM approach uses broad-based primers or ligated adapter sequences to universally amplify all nucleic acid molecules in a heterogeneous sample, which have been partitioned, as in digital PCR. Extensive assay optimization enables direct sequence identification by algorithm-based matching of melt curve shape and Tm to a database of known sequence-specific melt curves. We show that single-molecule detection and single nucleotide sensitivity is possible. The feasibility and utility of U-dHRM is demonstrated through detection of bacteria associated with polymicrobial blood infection and microRNAs (miRNAs) associated with host response to infection. U-dHRM using broad-based 16S rRNA gene primers demonstrates universal single cell detection of bacterial pathogens, even in the presence of larger amounts of contaminating bacteria; U-dHRM using universally adapted Lethal-7 miRNAs in a heterogeneous mixture showcases the single copy sensitivity and single nucleotide specificity of this approach.