QC-GN2oMS2: a Graph Neural Net for High Resolution Mass Spectra Prediction.

Predicting the mass spectrum of a molecular ion is often accomplished via three generalized approaches: rules-based methods for bond breaking, deep learning, or quantum chemical (QC) modeling. Rules-based approaches are often limited by the conditions for different chemical subspaces and perform poorly under chemical regimes with few defined rules. QC modeling is theoretically robust but requires significant amounts of computational time to produce a spectrum for a given target. Among deep learning techniques, graph neural networks (GNNs) have performed better than previous work with fingerprint-based neural networks in mass spectra prediction. To explore this technique further, we investigate the effects of including quantum chemically derived information as edge features in the GNN to increase predictive accuracy. The models we investigated include categorical bond order, bond force constants derived from extended tight-binding (xTB) quantum chemistry, and acyclic bond dissociation energies. We evaluated these models against a control GNN with no edge features in the input graphs. Bond dissociation enthalpies yielded the best improvement with a cosine similarity score of 0.462 relative to the baseline model (0.437). In this work we also apply dynamic graph attention which improves performance on benchmark problems and supports the inclusion of edge features. Between implementations, we investigate the nature of the molecular embedding for spectra prediction and discuss the recognition of fragment topographies in distinct chemistries for further development in tandem mass spectrometry prediction.

A Methodology for Evaluating Operator Usage of Machine Learning Recommendations for Power Grid Contingency Analysis.

This work presents the application of a methodology to measure domain expert trust and workload, elicit feedback, and understand the technological usability and impact when a machine learning assistant is introduced into contingency analysis for real-time power grid simulation. The goal of this framework is to rapidly collect and analyze a broad variety of human factors data in order to accelerate the development and evaluation loop for deploying machine learning applications. We describe our methodology and analysis, and we discuss insights gained from a pilot participant about the current usability state of an early technology readiness level (TRL) artificial neural network (ANN) recommender.

Augmentation of MS/MS Libraries with Spectral Interpolation for Improved Identification.

Tandem mass spectrometry (MS/MS) is a primary tool for the identification of small molecules and metabolites where resultant spectra are most commonly identified by matching them with spectra in MS/MS reference libraries. The high degree of variability in MS/MS spectrum acquisition techniques and parameters creates a significant challenge for building standardized reference libraries. Here we present a method to improve the usefulness of existing MS/MS libraries by augmenting available experimental spectra data sets with statistically interpolated spectra at unreported collision energies. We find that highly accurate spectral approximations can be interpolated from as few as three experimental spectra and that the interpolated spectra will be consistent with true spectra gathered from the same instrument as the experimental spectra. Supplementing existing spectral databases with interpolated spectra yields consistent improvements to identification accuracy on a range of instruments and precursor types. Applying this method yields significant improvements (∼10% more spectra correctly identified) on large data sets (2000-10 000 spectra), indicating this is a quick yet adept tool for improving spectral matching in situations where available reference libraries are not yet sufficient. We also find improvements of matching spectra across instrument types (between an Agilent Q-TOF and an Orbitrap Elite), at high collision energies (50-90 eV), and with smaller data sets available through MassBank.

A mathematical model for a biphasic DNA amplification reaction.

Isothermal DNA amplification reactions are a prevalent tool with many applications, ranging from analyte detection to DNA circuits. Exponential amplification reaction (EXPAR) is a popular isothermal DNA amplification method that exponentially amplifies short DNA oligonucleotides. A recent modification of this technique using an energetically stable looped template with palindromic binding regions demonstrated unexpected biphasic amplification and much higher DNA yield than EXPAR. This ultrasensitive DNA amplification reaction (UDAR) shows high-gain, switch-like DNA output from low concentrations of DNA input. Here we present the first mathematical model of UDAR based on four reaction mechanisms and show the model can reproduce the experimentally observed biphasic behaviour. Furthermore, we show that three of these mechanisms are necessary to reproduce biphasic experimental results. The reaction mechanisms are (i) positively cooperative multistep binding spurred by two trigger binding sites on the template; (ii) gradual template deactivation; (iii) recycling of deactivated templates into active templates; and (iv) polymerase sequestration. These mechanisms can potentially illuminate the behaviour of EXPAR as well as other nucleic acid amplification reactions.