Statistical Models for the Analysis and Design of Digital Polymerase Chain Reaction (dPCR) Experiments

Inferring transmission fitness advantage of SARS-CoV-2 variants of concern from wastewater samples using digital PCR, Switzerland, December 2020 through March 2021

BackgroundThroughout the COVID-19 pandemic, SARS-CoV-2 genetic variants of concern (VOCs) have repeatedly and independently arisen. VOCs are characterised by increased transmissibility, increased virulence or reduced neutralisation by antibodies obtained from prior infection or vaccination. Tracking the introduction and transmission of VOCs relies on sequencing, typically whole genome sequencing of clinical samples. Wastewater surveillance is increasingly used to track the introduction and spread of SARS-CoV-2 variants through sequencing approaches.AimHere, we adapt and apply a rapid, high-throughput method for detection and quantification of the relative frequency of two deletions characteristic of the Alpha, Beta, and Gamma VOCs in wastewater.MethodsWe developed drop-off RT-dPCR assays and an associated statistical approach implemented in the R package WWdPCR to analyse temporal dynamics of SARS-CoV-2 signature mutations (spike Δ69-70 and ORF1a Δ3675-3677) in wastewater and quantify transmission fitness advantage of the Alpha VOC.ResultsBased on analysis of Zurich wastewater samples, the estimated transmission fitness advantage of SARS-CoV-2 Alpha based on the spike Δ69-70 was 0.34 (95% confidence interval (CI): 0.30-0.39) and based on ORF1a Δ3675-3677 was 0.53 (95% CI: 0.49-0.57), aligning with the transmission fitness advantage of Alpha estimated by clinical sample sequencing in the surrounding canton of 0.49 (95% CI: 0.38-0.61).ConclusionDigital PCR assays targeting signature mutations in wastewater offer near real-time monitoring of SARS-CoV-2 VOCs and potentially earlier detection and inference on transmission fitness advantage than clinical sequencing.

Importance of sample input volume for accurate SARS-CoV-2 qPCR testing

Nucleic acid testing is the most widely used detection method for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the causative agent of the coronavirus disease 2019 (COVID-19) pandemic. Currently, a number of COVID-19 real-time quantitative reverse transcription PCR (qPCR) kits with high sensitivity and specificity are available for SARS-CoV-2 testing. However, these qPCR assays are not always reliable in detecting low viral load samples (Ct-value ≥ 35), resulting in inconclusive or false-negative results. Here, we used a Poisson distribution to illustrate the inconsistent performance of qPCR tests in detecting low viral load samples. From this, we concluded that the false-negative outcomes resulted from the random occurrences of sampling zero target molecules in a single test, and the probability to sample zero target molecules in one test decreased significantly with increasing purified RNA or initial sample input volume. At a given RNA concentration of 0.5 copy/μL, the probability of sampling zero RNA molecules decreased from 36.79% to close to 0.67% after increasing the RNA input volume from 2 to 10 μL. A SARS-CoV-2 qPCR assay with an LOD of 300 copies/mL was used to validate the improved consistency of the qPCR tests. We found that the false-negative qPCR results of clinical COVID-19 samples with a Ct ≥ 35 decreased by 50% after increasing the input of purified RNA from 2 to 10 μL. The consistency, accuracy, and robustness of nucleic acid testing for SARS-CoV-2 samples with low viral loads can be improved by increasing the sample input volume.

Optimizing the performance of digital loop-mediated isothermal amplification

Digital loop-mediated isothermal amplification (dLAMP) is attractive for the detection of nucleic acid due to its superior characteristics including isothermal amplification, absolute quantification, and single-molecule sensitivity. However, dLAMP suffers from the inaccurate quantification caused by low digital efficiency, which means only part of loaded template molecules could be amplified. We here developed a prehybridization-induced enhancement (PIE) strategy which could improve digital efficiency about 2-40 times without any new primer or additional operation. This work provides new insight into understanding the reaction dynamic of dLAMP. The PIE strategy could be applied to the other digital amplification methods.

Validation of Preamplification to Improve Quantification of Cytomegalovirus DNA Using Droplet Digital Polymerase Chain Reaction

Subclinical cytomegalovirus (CMV) replication is associated with strong cellular immune response and chronic inflammation, which could contribute to aging-related conditions such as cardiovascular disease and frailty. However, because of very low levels of CMV DNA present in people with chronic CMV infection, it has been difficult to explore the virologic and immunologic mechanisms of chronic low-level CMV infection and a sensitive method to monitor CMV replication is needed. Droplet digital PCR (ddPCR) has been shown to have higher precision and reproducibility than real-time quantitative PCR (qPCR) in quantifying low levels of CMV DNA, but it is not always sensitive enough for this purpose. Through rigorous validation experiments, we demonstrated that sensitivity and precision of quantification of very low levels of CMV DNA by ddPCR can be significantly increased by preamplification of samples with 10-20 cycles of conventional PCR, especially when testing CMV DNA in the presence of cellular DNA. With preamplification, we could reliably quantify down to two copies of CMV DNA, as opposed to five copies without preamplification. Further studies are needed to determine if ddPCR with preamplification can facilitate mechanistic studies of the characteristics and consequences of chronic CMV infection in aging adults.

A density-watershed algorithm (DWA) method for robust, accurate and automatic classification of dual-fluorescence and four-cluster droplet digital PCR data

Droplet digital PCR (ddPCR) is a single-molecule amplification technology with broad applications in precision medicine and clinical diagnosis. Dual-fluorescence and four-cluster ddPCR (two/four-ddPCR) assay is an effective way to quantify copy numbers. Currently, two/four-ddPCR data are usually classified with manual thresholds. For clinical applications, automatic and accurate methods are required to avoid subjectivity in diagnosis. Although there are some automatic classification algorithms, their accuracy and robustness still need to be improved to meet the needs of clinical diagnosis. Therefore, a new method is in high demand to automatically classify two/four-ddPCR data in an accurate and robust way. Here, a novel density-watershed algorithm (DWA) method was developed for the accurate, automatic and unsupervised classification of two/four-ddPCR data. First, data gridding was applied to a scatter plot of the fluorescence signal intensity to calculate data densities. Based on the data densities, the watershed algorithm was used to divide the gridded scatter plot into isolated regions automatically. Next, an optimal cluster pattern was determined based on these isolated regions, and excess regions were merged. Finally, the two/four-ddPCR data were classified based on the merged regions, and DNA template copy numbers were calculated accordingly. Using the DWA method for the quantification of both wild types and mutants of epidermal growth factor receptor (EGFR) L858R and T790M, the classification results were highly consistent with expectations, and significantly better than commonly-used automatic algorithms for now. The computed template copy numbers scaled proportionally to the relative concentration of input templates (r2 > 0.998) in four orders of magnitude with a good reproducibility, and achieved a limit of detection over 40 times lower than the commonly-used automatic algorithms. Furthermore, the DWA method was validated on 254 clinical DNA samples derived from frozen tissues, formalin-fixed paraffin-embedded tissues and peripheral blood. In most cases, the DWA method derived accurate and valid classification results. This highly effective DWA method may be widely used for automatically classifying two/four-ddPCR data, and it will greatly promote the application of ddPCR in clinical diagnosis.

Environmental DNA sampling reveals high occupancy rates of invasive Burmese pythons at wading bird breeding aggregations in the central Everglades

The Burmese python (Python bivittatus) is now established as a breeding population throughout south Florida, USA. However, the extent of the invasion, and the ecological impacts of this novel apex predator on animal communities are incompletely known, in large part because Burmese pythons (hereafter “pythons”) are extremely cryptic and there has been no efficient way to detect them. Pythons are recently confirmed nest predators of long-legged wading bird breeding colonies (orders Ciconiiformes and Pelecaniformes). Pythons can consume large quantities of prey and may not be recognized as predators by wading birds, therefore they could be a particular threat to colonies. To quantify python occupancy rates at tree islands where wading birds breed, we utilized environmental DNA (eDNA) analysis—a genetic tool which detects shed DNA in water samples and provides high detection probabilities. We fitted multi-scale Bayesian occupancy models to test the prediction that pythons occupy islands with wading bird colonies at higher rates compared to representative control islands containing no breeding birds. Our results suggest that pythons are widely distributed across the central Everglades in proximity to active wading bird colonies. In support of our prediction that pythons are attracted to colonies, site-level python eDNA occupancy rates were higher at wading bird colonies (ψ = 0.88, 95% credible interval [0.59–1.00]) than at the control islands (ψ = 0.42 [0.16–0.80]) in April through June (n = 15 colony-control pairs). We found our water temperature proxy (time of day) to be informative of detection probability, in accordance with other studies demonstrating an effect of temperature on eDNA degradation in occupied samples. Individual sample concentrations ranged from 0.26 to 38.29 copies/μL and we generally detected higher concentrations of python eDNA in colony sites. Continued monitoring of wading bird colonies is warranted to determine the effect pythons are having on populations and investigate putative management activities.

Self-digitization chip for quantitative detection of human papillomavirus gene using digital LAMP

Digital nucleic acid amplification and detection methods provide excellent sensitivity and specificity and allow absolute quantification of target nucleic acids. Isothermal methods such as digital loop-mediated isothermal amplification (digital LAMP) have potential for use in rapid disease diagnosis in low-resource settings due to their speed and lack of thermal cycling. We previously developed a self-digitization (SD) chip, a simple microfluidics device that automatically digitizes a sample into an array of nanoliter wells, for use in digital LAMP. In this work, we improve the SD chip design to increase sample loading efficiency, speed, and completeness, and test a range of well volumes and numbers. We demonstrate the diagnostic capability of this platform by applying it to quantifying human papillomavirus 18 gene.

Inhibition mechanisms of hemoglobin, immunoglobulin G, and whole blood in digital and real-time PCR

Blood samples are widely used for PCR-based DNA analysis in fields such as diagnosis of infectious diseases, cancer diagnostics, and forensic genetics. In this study, the mechanisms behind blood-induced PCR inhibition were evaluated by use of whole blood as well as known PCR-inhibitory molecules in both digital PCR and real-time PCR. Also, electrophoretic mobility shift assay was applied to investigate interactions between inhibitory proteins and DNA, and isothermal titration calorimetry was used to directly measure effects on DNA polymerase activity. Whole blood caused a decrease in the number of positive digital PCR reactions, lowered amplification efficiency, and caused severe quenching of the fluorescence of the passive reference dye 6-carboxy-X-rhodamine as well as the double-stranded DNA binding dye EvaGreen. Immunoglobulin G was found to bind to single-stranded genomic DNA, leading to increased quantification cycle values. Hemoglobin affected the DNA polymerase activity and thus lowered the amplification efficiency. Hemoglobin and hematin were shown to be the molecules in blood responsible for the fluorescence quenching. In conclusion, hemoglobin and immunoglobulin G are the two major PCR inhibitors in blood, where the first affects amplification through a direct effect on the DNA polymerase activity and quenches the fluorescence of free dye molecules, and the latter binds to single-stranded genomic DNA, hindering DNA polymerization in the first few PCR cycles. Graphical abstract PCR inhibition mechanisms of hemoglobin and immunoglobulin G (IgG). Cq quantification cycle, dsDNA double-stranded DNA, ssDNA single-stranded DNA.

Accurate Digital Polymerase Chain Reaction Quantification of Challenging Samples Applying Inhibitor-Tolerant DNA Polymerases

Digital PCR (dPCR) enables absolute quantification of nucleic acids by partitioning of the sample into hundreds or thousands of minute reactions. By assuming a Poisson distribution for the number of DNA fragments present in each chamber, the DNA concentration is determined without the need for a standard curve. However, when analyzing nucleic acids from complex matrixes such as soil and blood, the dPCR quantification can be biased due to the presence of inhibitory compounds. In this study, we evaluated the impact of varying the DNA polymerase in chamber-based dPCR for both pure and impure samples using the common PCR inhibitor humic acid (HA) as a model. We compared the TaqMan Universal PCR Master Mix with two alternative DNA polymerases: ExTaq HS and Immolase. By using Bayesian modeling, we show that there is no difference among the tested DNA polymerases in terms of accuracy of absolute quantification for pure template samples, i.e., without HA present. For samples containing HA, there were great differences in performance: the TaqMan Universal PCR Master Mix failed to correctly quantify DNA with more than 13 pg/nL HA, whereas Immolase (1 U) could handle up to 375 pg/nL HA. Furthermore, we found that BSA had a moderate positive effect for the TaqMan Universal PCR Master Mix, enabling accurate quantification for 25 pg/nL HA. Increasing the amount of DNA polymerase from 1 to 5 U had a strong effect for ExTaq HS, elevating HA-tolerance four times. We also show that the average Cq values of positive reactions may be used as a measure of inhibition effects, e.g., to determine whether or not a dPCR quantification result is reliable. The statistical models developed to objectively analyze the data may also be applied in quality control. We conclude that the choice of DNA polymerase in dPCR is crucial for the accuracy of quantification when analyzing challenging samples.

A Robust Protocol for Using Multiplexed Droplet Digital PCR to Quantify Somatic Copy Number Alterations in Clinical Tissue Specimens

The ability of droplet digital PCR (ddPCR) to accurately determine the concentrations of amplifiable targets makes it a promising platform for measuring copy number alterations (CNAs) in genomic biomarkers. However, its application to clinical samples, particularly formalin-fixed paraffin-embedded specimens, will require strategies to reliably determine CNAs in DNA of limited quantity and quality. When applied to cancerous tissue, those methods must also account for global genetic instability and the associated probability that the abundance(s) of one or more chosen reference loci do not represent the average ploidy of cells comprising the specimen. Here we present an experimental design strategy and associated data analysis tool that enables accurate determination of CNAs in a panel of biomarkers using multiplexed ddPCR. The method includes strategies to optimize primer and probes design to cleanly segregate droplets in the data output from reaction wells amplifying multiple independent templates, and to correct for bias from artifacts such as DNA fragmentation. We demonstrate how a panel of reference loci can be used to determine a stable CNA-neutral benchmark. These innovations, when taken together, provide a comprehensive strategy that can be used to reliably detect biomarker CNAs in DNA extracted from either frozen or FFPE tissue biopsies.

Methods for comparing multiple digital PCR experiments

The estimated mean copy per partition (λ) is the essential information from a digital PCR (dPCR) experiment because λ can be used to calculate the target concentration in a sample. However, little information is available how to statistically compare dPCR runs of multiple runs or reduplicates. The comparison of λ values from several runs is a multiple comparison problem, which can be solved using the binary structure of dPCR data. We propose and evaluate two novel methods based on Generalized Linear Models (GLM) and Multiple Ratio Tests (MRT) for comparison of digital PCR experiments. We enriched our MRT framework with computation of simultaneous confidence intervals suitable for comparing multiple dPCR runs. The evaluation of both statistical methods support that MRT is faster and more robust for dPCR experiments performed in large scale. Our theoretical results were confirmed by the analysis of dPCR measurements of dilution series.

Both methods were implemented in the dpcR package (v. 0.2) for the open source R statistical computing environment.

Flexible analysis of digital PCR experiments using generalized linear mixed models

The use of digital PCR for quantification of nucleic acids is rapidly growing. A major drawback remains the lack of flexible data analysis tools. Published analysis approaches are either tailored to specific problem settings or fail to take into account sources of variability. We propose the generalized linear mixed models framework as a flexible tool for analyzing a wide range of experiments. We also introduce a method for estimating reference gene stability to improve accuracy and precision of copy number and relative expression estimates. We demonstrate the usefulness of the methodology on a complex experimental setup.

Designing and interpretation of digital assays: Concentration of target in the sample and in the source of sample

We explain how to design classic digital assays, comprising identical partitions, in order to obtain the required precision of the estimate within a defined range of concentrations. The design, including the number and volume of partitions, depends significantly on whether the assay is to assess the concentration of the target analyte in the sample or in the source of the sample (e.g. a patient body) with a given precision. We also show how to translate the result referring to the concentration in the sample into the concentration in the source of the sample, including the significant change in the breath of the confidence intervals.