Stephen Winters-Hilt

Rapid discrimination among individual DNA hairpin molecules at single-nucleotide resolution using an ion channel

RNA and DNA strands produce ionic current signatures when driven through an α-hemolysin channel by an applied voltage. Here we combine this nanopore detector with a support vector machine (SVM) to analyze DNA hairpin molecules on the millisecond time scale. Measurable properties include duplex stem length, base pair mismatches, and loop length. This nanopore instrument can discriminate between individual DNA hairpins that differ by one base pair or by one nucleotide.

Highly Accurate Classification of Watson-Crick Basepairs on Termini of Single DNA Molecules

We introduce a computational method for classification of individual DNA molecules measured by an alpha-hemolysin channel detector. We show classification with better than 99% accuracy for DNA hairpin molecules that differ only in their terminal Watson-Crick basepairs. Signal classification was done in silico to establish performance metrics (i.e., where train and test data were of known type, via single-species data files). It was then performed in solution to assay real mixtures of DNA hairpins. Hidden Markov Models (HMMs) were used with Expectation/Maximization for denoising and for associating a feature vector with the ionic current blockade of the DNA molecule. Support Vector Machines (SVMs) were used as discriminators, and were the focus of off-line training. A multiclass SVM architecture was designed to place less discriminatory load on weaker discriminators, and novel SVM kernels were used to boost discrimination strength. The tuning on HMMs and SVMs enabled biophysical analysis of the captured molecule states and state transitions; structure revealed in the biophysical analysis was used for better feature selection.

Implementing EM and Viterbi algorithms for Hidden Markov Model in linear memory

BackgroundThe Baum-Welch learning procedure for Hidden Markov Models (HMMs) provides a powerful tool for tailoring HMM topologies to data for use in knowledge discovery and clustering. A linear memory procedure recently proposed by Miklós, I. and Meyer, I.M. describes a memory sparse version of the Baum-Welch algorithm with modifications to the original probabilistic table topologies to make memory use independent of sequence length (and linearly dependent on state number). The original description of the technique has some errors that we amend. We then compare the corrected implementation on a variety of data sets with conventional and checkpointing implementations.ResultsWe provide a correct recurrence relation for the emission parameter estimate and extend it to parameter estimates of the Normal distribution. To accelerate estimation of the prior state probabilities, and decrease memory use, we reverse the originally proposed forward sweep. We describe different scaling strategies necessary in all real implementations of the algorithm to prevent underflow. In this paper we also describe our approach to a linear memory implementation of the Viterbi decoding algorithm (with linearity in the sequence length, while memory use is approximately independent of state number). We demonstrate the use of the linear memory implementation on an extended Duration Hidden Markov Model (DHMM) and on an HMM with a spike detection topology. Comparing the various implementations of the Baum-Welch procedure we find that the checkpointing algorithm produces the best overall tradeoff between memory use and speed. In cases where sequence length is very large (for Baum-Welch), or state number is very large (for Viterbi), the linear memory methods outlined may offer some utility.ConclusionOur performance-optimized Java implementations of Baum-Welch algorithm are available at http://logos.cs.uno.edu/~achurban. The described method and implementations will aid sequence alignment, gene structure prediction, HMM profile training, nanopore ionic flow blockades analysis and many other domains that require efficient HMM training with EM.

Duration learning for analysis of nanopore ionic current blockades

BackgroundIonic current blockade signal processing, for use in nanopore detection, offers a promising new way to analyze single molecule properties, with potential implications for DNA sequencing. The alpha-Hemolysin transmembrane channel interacts with a translocating molecule in a nontrivial way, frequently evidenced by a complex ionic flow blockade pattern. Typically, recorded current blockade signals have several levels of blockade, with various durations, all obeying a fixed statistical profile for a given molecule. Hidden Markov Model (HMM) based duration learning experiments on artificial two-level Gaussian blockade signals helped us to identify proper modeling framework. We then apply our framework to the real multi-level DNA hairpin blockade signal.ResultsThe identified upper level blockade state is observed with durations that are geometrically distributed (consistent with an a physical decay process for remaining in any given state). We show that mixture of convolution chains of geometrically distributed states is better for presenting multimodal long-tailed duration phenomena. Based on learned HMM profiles we are able to classify 9 base-pair DNA hairpins with accuracy up to 99.5% on signals from same-day experiments.ConclusionWe have demonstrated several implementations for de novo estimation of duration distribution probability density function with HMM framework and applied our model topology to the real data. The proposed design could be handy in molecular analysis based on nanopore current blockade signal.

Nanopore Detector based analysis of single-molecule conformational kinetics and binding interactions

BackgroundA Nanopore Detector provides a means to transduce single molecule events into observable channel current changes. Nanopore-based detection can report directly, or indirectly, on single molecule kinetics. The nanopore-based detector can directly measure molecular characteristics in terms of the blockade properties of individual molecules – this is possible due to the kinetic information that is embedded in the blockade measurements, where the adsorption-desorption history of the molecule to the surrounding channel, and the configurational changes in the molecule itself, imprint on the ionic flow through the channel. This rich source of information offers prospects for DNA sequencing and single nucleotide polymorphism (SNP) analysis. A nanopore-based detector can also measure molecular characteristics indirectly, by using a reporter molecule that binds to certain molecules, with subsequent distinctive blockade by the bound-molecule complex.ResultsIt is hypothesized that reaction histories of individual molecules can be observed on model DNA/DNA, DNA/Protein, and Protein/Protein systems. Preliminary results are all consistent with this hypothesis. Nanopore detection capabilities are also described for highly discriminatory biosensing, binding strength characterization, and rapid immunological screening.ConclusionIn essence, the heart of chemistry is now accessible to a new, single-molecule, observation method that can track both external molecular binding states, and internal conformation states.

Clustering ionic flow blockade toggles with a Mixture of HMMs

BackgroundIonic current blockade signal processing, for use in nanopore detection, offers a promising new way to analyze single molecule properties with potential implications for DNA sequencing. The α-Hemolysin transmembrane channel interacts with a translocating molecule in a nontrivial way, frequently evidenced by a complex ionic flow blockade pattern with readily distinguishable modes of toggling. Effective processing of such signals requires developing machine learning methods capable of learning the various blockade modes for classification and knowledge discovery purposes. Here we propose a method aimed to improve our stochastic analysis capabilities to better understand the discriminatory capabilities of the observed the nanopore channel interactions with analyte.ResultsWe tailored our memory-sparse distributed implementation of a Mixture of Hidden Markov Models (MHMMs) to the problem of channel current blockade clustering and associated analyte classification. By using probabilistic fully connected HMM profiles as mixture components we were able to cluster the various 9 base-pair hairpin channel blockades. We obtained very high Maximum a Posteriori (MAP) classification with a mixture of 12 different channel blockade profiles, each with 4 levels, a configuration that can be computed with sufficient speed for real-time experimental feedback. MAP classification performance depends on several factors such as the number of mixture components, the number of levels in each profile, and the duration of a channel blockade event. We distribute Baum-Welch Expectation Maximization (EM) algorithms running on our model in two ways. A distributed implementation of the MHMM data processing accelerates data clustering efforts. The second, simultanteous, strategy uses an EM checkpointing algorithm to lower the memory use and efficiently distribute the bulk of EM processing in processing large data sequences (such as for the progressive sums used in the HMM parameter estimates).ConclusionThe proposed distributed MHMM method has many appealing properties, such as precise classification of analyte in real-time scenarios, and the ability to incorporate new domain knowledge into a flexible, easily distributable, architecture. The distributed HMM provides a feature extraction that is equivalent to that of the sequential HMM with a speedup factor approximately equal to the number of independent CPUs operating on the data. The MHMM topology learns clusters existing within data samples via distributed HMM EM learning. A Java implementation of the MHMM algorithm is available at http://logos.cs.uno.edu/~achurban.

Highly Accurate Real‐Time Classification of Channel‐Captured DNA Termini

A computational method is briefly described for classification of individual DNA molecules measured by an α‐hemolysin channel detector. Classification is performed with better than 99% accuracy for DNA hairpin molecules that differ only in their terminal Watson‐Crick base pairs. Signal classification was initially done on synthetic data streams, where sampling on real mixtures of hairpins was modeled in order to establish performance metrics (i.e., where train and test data were of known type, via single‐species data files). Signal classification was then performed on observations from real mixtures of DNA hairpins. Hidden Markov Models (HMMs) were used with Expectation/Maximization for de‐noising and for associating a feature vector with the ionic current blockade of the DNA molecule. Support Vector Machines (SVMs) were used as discriminators, and were the focus of off‐line training. A multi‐class SVM architecture was designed to place less discriminatory load on weaker discriminators, and novel SVM kern...

Nanopore detection using channel current cheminformatics

A novel detector is used for analysis of single DNA molecules. The detector is based on current blockade measurements through a single, nanometer-scale, α-hemolysin ion channel. The biologically based alpha-hemolysin channel self-assembles in lipid bilayers, permitting highly reproducible experiments with Angstrom resolution. In previous work the spectrum of dsDNA blockade states could be explained in terms of the dsDNA-protein binding kinetics, and dsDNA terminus fraying (bond dissociation) kinetics. Results presented here strengthen the hypothesis that conformational dynamics can be observed as well, when the channel-captured dsDNA end is in an unbound state. Feature discovery methods: include a time-domain finite state automaton (FSA), a wavelet domain FSA, and a Hidden Markov Model (HMM). Classifier feature extraction methods: includes a time-domain FSA for signal acquisition and a generalized HMM with EM for features extraction. Classification method: Support Vector Machines (SVMs) are used with novel kernel designs. Kinetic feature extraction tool: a time-domain FSA projects current observations to a (small) set of blockade states. Those states are provided by the generalized HMM analysis. Noise sources limit the resolution of the nanopore device, and its multiclass scaling capabilities, and this is discussed in the context of ongoing refinements to the device.

The NTD Nanoscope: potential applications and implementations

BackgroundNanopore transduction detection (NTD) offers prospects for a number of highly sensitive and discriminative applications, including: (i) single nucleotide polymorphism (SNP) detection; (ii) targeted DNA re-sequencing; (iii) protein isoform assaying; and (iv) biosensing via antibody or aptamer coupled molecules. Nanopore event transduction involves single-molecule biophysics, engineered information flows, and nanopore cheminformatics. The NTD Nanoscope has seen limited use in the scientific community, however, due to lack of information about potential applications, and lack of availability for the device itself. Meta Logos Inc. is developing both pre-packaged device platforms and component-level (unassembled) kit platforms (the latter described here). In both cases a lipid bi-layer workstation is first established, then augmentations and operational protocols are provided to have a nanopore transduction detector. In this paper we provide an overview of the NTD Nanoscope applications and implementations. The NTD Nanoscope Kit, in particular, is a component-level reproduction of the standard NTD device used in previous research papers.ResultsThe NTD Nanoscope method is shown to functionalize a single nanopore with a channel current modulator that is designed to transduce events, such as binding to a specific target. To expedite set-up in new lab settings, the calibration and troubleshooting for the NTD Nanoscope kit components and signal processing software, the NTD Nanoscope Kit, is designed to include a set of test buffers and control molecules based on experiments described in previous NTD papers (the model systems briefly described in what follows). The description of the Server-interfacing for advanced signal processing support is also briefly mentioned.ConclusionsSNP assaying, SNP discovery, DNA sequencing and RNA-seq methods are typically limited by the accuracy of the error rate of the enzymes involved, such as methods involving the polymerase chain reaction (PCR) enzyme. The NTD Nanoscope offers a means to obtain higher accuracy as it is a single-molecule method that does not inherently involve use of enzymes, using a functionalized nanopore instead.

A Hidden Markov Model With Binned Duration Algorithm

The hidden Markov model with duration (HMMD) is critically important when the distributions on state intervals deviate significantly from the geometric distribution, such as for multimodal distributions and heavy-tailed distributions. Heavy-tailed distributions, in particular, are widespread in describing phenomena across the sciences, where the log-normal, students-T, and Pareto distributions are heavy-tailed distributions that are almost as common as the normal and geometric distributions in descriptions of physical phenomena or man-made phenomena. The standard hidden Markov model (HMM) constrains state occupancy durations to be geometrically distributed, while HMMD avoids this limitation, but at significant computational expense. We propose a new algorithm, hidden Markov model with binned duration, whose result shows no loss of accuracy compared to the HMMD decoding performance and a computational expense that only differs from the much simpler and faster HMM decoding by a constant factor.