Artificial intelligence systems for molecular spectral analysis

A brief survey is given of the most recent publications on development of artificial-intelligence systems for molecular spectral analysis. A new approach to solution of the problems of qualitative molecular spectral analysis is based on an applied logical calculus developed by the authors for fuzzy predicates. It is suggested that spectral-structural knowledge should be specified in the language of fuzzy predicates, and mechanical theorem-proving procedures used for solving qualitative problems of spectral analysis, the initial information being considered as a set of axioms. System-oriented matters are given consideration. The formalism suggested is a basis for the development of an artificial-intelligence dialogue system capable of solving various problems in molecular spectral analysis while maintaining a dialogue with a research worker using a professionally-restricted natural language.     

Swarm intelligence based wavelet coefficient feature selection for mass spectral classification: An application to proteomics data

This paper introduces the ant colony algorithm, a novel swarm intelligence based optimization method, to select appropriate wavelet coefficients from mass spectral data as a new feature selection method for ovarian cancer diagnostics. By determining the proper parameters for the ant colony algorithm (ACA) based searching algorithm, we perform the feature searching process for 100 times with the number of selected features fixed at 5. The results of this study show: (1) the classification accuracy based on the five selected wavelet coefficients can reach up to 100% for all the training, validating and independent testing sets; (2) the eight most popular selected wavelet coefficients of the 100 runs can provide 100% accuracy for the training set, 100% accuracy for the validating set, and 98.8% accuracy for the independent testing set, which suggests the robustness and accuracy of the proposed feature selection method; and (3) the mass spectral data corresponding to the eight popular wavelet coefficients can be located by reverse wavelet transformation and these located mass spectral data still maintain high classification accuracies (100% for the training set, 97.6% for the validating set, and 98.8% for the testing set) and also provide sufficient physical and medical meaning for future ovarian cancer mechanism studies. Furthermore, the corresponding mass spectral data (potential biomarkers) are in good agreement with other studies which have used the same sample set. Together these results suggest this feature extraction strategy will benefit the development of intelligent and real-time spectroscopy instrumentation based diagnosis and monitoring systems.     

Three-dimensional mass spectral imaging of HeLa-M cells--sample preparation, data interpretation and visualisation

Time-of-flight secondary ion mass spectrometry (ToFSIMS) is being applied increasingly to the study of biological systems where the chemical specificity of mass spectrometry and the high lateral resolution imaging capabilities can be exploited. Here we report a comparison of two cell sample preparation methods and demonstrate how they influence the outcome of the ToFSIMS analysis for three-dimensional (3D) imaging of biological cells using our novel buncher-ToF instrument (J105 3D Chemical Imager) equipped with a C(60) primary ion beam. Cells were analysed fixed and freeze-dried and non-fixed, frozen-hydrated. It is concluded that maintaining the cells in a non-fixed frozen-hydrated state during the analysis helps reduce chemical redistribution, producing cleaner spectra and improved chemical contrast in both 2D and 3D imaging. Insights into data interpretation are included and we present methods for 3D reconstruction of the data using multivariate analysis techniques.     

Energetic data as a guide to the interpretation of mass spectral features of dicarbonyl diketonates of Rh and Ir

The relative intensities and appearance potentials of the principal ions formed from a series of rhodium and iridium dicarbonyl diketonates were measured by mass spectrometry. These compounds show the usual CO losses generally found in the fragmentation of metal carbonyls. Additionally several fragmentation peculiarities characteristic of metal acetylacetonates were found. Energetic data were used to interpret some aspects of the cracking patterns.     

Artificial intelligence and the competitiveness of perturbative approaches

Due to their many past shortcomings perturbation techniques have been seldom used in theoretical chemistry. However, the introduction of operator- and superoperator-based techniques has represented a real breakthrough in their usefulness as a powerful tool of calculation. We analyze here how an artificial intelligence (AI) approach to their automation can save the obstacle of their cumbersome algebra and, by manipulating full analytic expressions with the help of a simple combinatorial analysis, represents an attractive alternative to more orthodox, purely numerical, approaches.     

The computer processing and interpretation of mass spectral information. VI—computing the isotopic spectrum of assumed composition

An interactive program for the computation of an isotopic spectrum of assumed composition whose empirical formula can be represented either as a set of chemical elements or in terms of structural fragments (super-elements) is described. The core of the program is a table of all isotopes of 82 elements of the periodic table and super-elements whose list is decided on by the user. The exact mass of isotopic peaks is determined by the program either from the position of the maximum value of the peak abundance or from the position of its centroid. The spectrum printout may be in one of several formats. The interaction with the program is very useful for operators of mass spectrometers in enabling them to estimate and verify assumptions about the elemental composition of spectrum peak groups.     

New interpretation of the AgL-series x-ray spectrum

Based on the de-excitation model of atomicL-shell vacancies, recent put forward by us, a new interpretation of Paratt's AgL x-ray spectrum has been successfully carried out, by using experimental and semiempiricalL ₁-hole decay data reported in recent years. An initialL-vacancy distributionN ₁:N ₂:N ₃=1:1.44:3.14 has been calculated from the silver spectrum. Present value of AgM-electron shakeoff probability due to anL-shell hole isP _M =0.034, which is much larger than theoretical calculations and semiempirical data.     

Artificial intelligence and machine learning for targeted energy storage solutions

With the application of machine learning to large-material data sets, models are being developed that allow us to better predict novel materials with designed properties. Advances in artificial intelligence and its subclasses, as well as compute infrastructure, are making it possible to rapidly compute material properties, to access time/length scales and chemical spaces beyond the current capabilities of density functional theory and to outperform humans in interpretation and characterization of the data. This review highlights the latest developments in the field with special interest to energy storage materials.     

MassBank: a public repository for sharing mass spectral data for life sciences

MassBank is the first public repository of mass spectra of small chemical compounds for life sciences (<3000 Da). The database contains 605 electron‐ionization mass spectrometry(EI‐MS), 137 fast atom bombardment MS and 9276 electrospray ionization (ESI)‐MSⁿ data of 2337 authentic compounds of metabolites, 11 545 EI‐MS and 834 other‐MS data of 10 286 volatile natural and synthetic compounds, and 3045 ESI‐MS² data of 679 synthetic drugs contributed by 16 research groups (January 2010). ESI‐MS² data were analyzed under nonstandardized, independent experimental conditions. MassBank is a distributed database. Each research group provides data from its own MassBank data servers distributed on the Internet. MassBank users can access either all of the MassBank data or a subset of the data by specifying one or more experimental conditions. In a spectral search to retrieve mass spectra similar to a query mass spectrum, the similarity score is calculated by a weighted cosine correlation in which weighting exponents on peak intensity and the mass‐to‐charge ratio are optimized to the ESI‐MS² data. MassBank also provides a merged spectrum for each compound prepared by merging the analyzed ESI‐MS² data on an identical compound under different collision‐induced dissociation conditions. Data merging has significantly improved the precision of the identification of a chemical compound by 21–23% at a similarity score of 0.6. Thus, MassBank is useful for the identification of chemical compounds and the publication of experimental data. Copyright © 2010 John Wiley & Sons, Ltd.     

Artificial intelligence techniques for modeling and optimization of the HDS process over a new graphene based catalyst

A Co-Mo/graphene oxide (GO) catalyst has been synthesized for the first time for application in a defined hydrodesulfurization (HDS) process to produce sulfur free naphtha. An intelligent model based upon the neural network technique has then been developed to estimate the total sulfur output of this process. Process operating variables include temperature, pressure, LHSV and H₂/feed volume ratio. The three-layer, feed-forward neural network developed consists of five neurons in a hidden layer, trained with Levenberg–Marquardt, back-propagation gradient algorithm. The predicted amount of residual total sulfur is in very good agreement with the corresponding experimental values revealing a correlation coefficient of greater than 0.99. In addition, a genetic algorithm (GA) has been employed to optimize values of total sulfur as well as reaction conditions.     

Applications of artificial intelligence for chemical inference—X : Intsum. A data interpretation and summary program applied to the collected mass spectra of estrogenic steroids

A method for systematic interpretation and summary of evidence found for all possible mass spectral fragmentations of a molecule or set of related molecules is described. The method is embodied in a computer program (INTSUM) which interprets, in terms of fragmentation processes, mass spectral data collected on known compounds. Utilizing high resolution mass spectra from 47 estrogenic steroids, the method is verified and new findings are discussed. Finally, the method is used to explore the fragmentations of equilenins and several acetate and benzoate ester derivatives.     

An interpretation of the N2 photoelectron spectrum

The inner-valence (20–36 eV) photoelectron spectrum of N₂ is interpreted by comparing with various spectra (such as absorption, N⁺ yield, fluorescence-yield, and core-level photoelectron spectra). This comparative study confirms the previous assignments on the bands at 25.3 and 29.0 eV and clarifies the controversial assignments on the bands at 32.6 and 35 eV (weak). Also, this study identifies for the first time a weak but sharp band at 33.3 eV.     

Vibrational interference effects in x-ray emission of a model water dimer: implications for the interpretation of the liquid spectrum

We apply the Kramers-Heisenberg formula to a model water dimer to discuss vibrational interference in the x-ray emission spectrum of the donor molecule for which the core-ionized potential energy surface is dissociative but bounded by the accepting molecule. A long core-hole lifetime leads to decay from Zundel-like, fully delocalized vibrational states in the intermediate potential without involvement of a specific dissociated component. Comparison is made to a model with an unbound intermediate state allowing dissociation to infinity which gives a sharp, fully dissociated feature, and a broad molecular peak at long core-hole life time. The implications of the vibrational interference effect on the liquid water spectrum are discussed and it is proposed that this mainly gives rise to an isotope-dependent asymmetrical broadening of the lone pair peak.     

Application of the combined CFD and swarm intelligence for optimization of baffles number in a mixer-settler

Process optimization in a mixer-settler is of great importance. Optimization algorithm of particle swarm optimization is one of the evolutionary algorithms to solve optimization problem which is used in many fields. In this study, the optimal condition is calculated in finite volume method in terms of the number of baffles, inlet velocity of fluid, and impeller speed in a mixer-settler with a specific dimension that can be extended to industrial dimensions using the PSO algorithm and the numerical solution of Navier-Stokes equations and k-ε standard.