Development of algorithms for automated elucidation of spectral feature/substructure relationships in tandem mass spectrometry

A pattern-recognition/artificial-intelligence program, referred to as MAPS (method for analyzing patterns in spectra), is described for the identification of relationships that exist between the presence of substructures in molecules and the characteristic features they produce in mass spectrometry (MS) and MS/MS data. The MAPS algorithm discovers these relationships by intelligent analysis of a data base of MS and MS/MS spectra. The relationships found are expressed as rules, which may then be used to identify characterized substructures in “unknowns”. No prior knowledge of fragmentation pathways or rearrangements is assumed in the rule-generation process. While MAPS currently uses MS and MS/MS data, the approach (and much of the software) is equally suited to multiple-stage mass spectrometric data.     

Computer-assisted prediction of pesticide substructure using mass spectra

Mass spectral classifiers of 16 substructures that are present in basic structures of pesticides have been investigated to assist pesticide residues analysis as well as screening of pesticide lead compounds. Mass spectral data are first transformed into 396 features, and then Genetic Algorithm-Partial Least Squares (GA-PLS) as a feature selection method and Support Vector Machine (SVM) as a validation method are implemented together to get an optimization feature set for each substructure. At last, a statistical method which is AdaBoost algorithm combined with Classification and Regression Tree (AdaBoost-CART) is trained to predict the 16 substructures presence/absence using the optimization mass spectral feature set. It is demonstrated that the optimum feature sets can be used to predict the 16 pesticide substructures presence/absence with mostly 85-100% in recognition success rate instead of the original 396 features.     

IEF pattern classification‐derived criteria for the identification of epoetin‐δ in urine

Epoetin‐δ (Dynepo? Shire Pharmaceuticals, Basing stoke, UK) is a synthetic form of erythropoietin (EPO) whose resemblance with endogenous EPO makes it hard to identify using the classical identification criteria. Urine samples collected from six healthy volunteers treated with epoetin‐δ injections and from a control population were immuno‐purified and analyzed with the usual IEF method. On the basis of the EPO profiles integration, a linear multivariate model was computed for discriminant analysis. For each sample, a pattern classification algorithm returned a bands distribution and intensity score (bands intensity score) saying how representative this sample is of one of the two classes, positive or negative. Effort profiles were also integrated in the model. The method yielded a good sensitivity versus specificity relation and was used to determine the detection window of the molecule following multiple injections. The bands intensity score, which can be generalized to epoetin‐α and epoetin‐β, is proposed as an alternative criterion and a supplementary evidence for the identification of EPO abuse.     

利用GC -MS结合移动窗口正交投影法及香原料数据库定性混配香精中的复杂成分

使用气相色谱-质谱( GC - MS)对含有表香干扰的混配香精(由5种精油和3种酊剂加入表香中混配而成)的复杂成分进行分析,利用移动窗口特征峰正交投影法,结合使用已建立的香原料物质数据库,快速判别了混配香精中的精油及酊剂等复杂成分,并详细研究了从复杂混配香精中剖析精油、酊剂等复杂成分的影响因素,从而初步提出一种从复杂混...     

Database clustering with a combination of fingerprint and maximum common substructure methods

We present an efficient method to cluster large chemical databases in a stepwise manner. Databases are first clustered with an extended exclusion sphere algorithm based on Tanimoto coefficients calculated from Daylight fingerprints. Substructures are then extracted from clusters by iterative application of a maximum common substructure algorithm. Clusters with common substructures are merged through a second application of an exclusion sphere algorithm. In a separate step, singletons are compared to cluster substructures and added to a cluster if similarity is sufficiently high. The method identifies tight clusters with conserved substructures and generates singletons only if structures are truly distinct from all other library members. The method has successfully been applied to identify the most frequently occurring scaffolds in databases, for the selection of analogues of screening hits and in the prioritization of chemical libraries offered by commercial vendors.     

Compensation for mobility inequalities between lanes computed from band signals in on-line fluorescence DNA sequencing.

Among on-line fluorescence DNA sequencing systems, the four-lane method exhibits the potential for reporting an erroneous sequence due to nonuniform mobility of the DNA fragments migrating among the four lanes. This error is manifest in phenomenon commonly called smiling. This paper presents a computational algorithm which compensates for the mobility inequalities between lanes using signal data obtained from the shorter DNA fragments forming the faster migrating bands. The program mainly consists of two routines: (i) calculation of calibration coefficients (mobility ratios between lanes), and (ii) examination of the coefficients by applying them to a later domain of the same signals. Both routines are connected with several feed-back branches for recalculation. Homology analysis of final sequences has shown that the accuracy rate is maximized with this algorithm and any ambiguous result can be assigned to the residual error inherent in the band identification method used.     

Computer based IR-identification of substructures

There are many approaches to rationalizing and speeding up the empirical interpretation of the infra-red (IR) spectrum of a complex molecule, in order to determine its chemical structure. One of these techniques is the identification of substructures by means of a matrix algorithm.     

Computer-aided interpretation of mass spectra. Information on substructural probabilities form stirs

The capability of the ‘Self-Training Interpretive and Retrieval System’ has been extended so that the compounds selected as providing the best match in each data class to the unknown mass spectrum are examined by the computer for the presence of each of 179 common substructural fragments. Stastical methods were used to evaluate the selectivity for identification of each substructure by each data class using a reference file of 18 806 spectra of different compounds. In tests using at least 373 unknown spectra for each of the substructures, with criteria that gave a mean reliability of 98.1%, the method averaged 49% recall, which corresponds to the identification of 2.55 substructures per unknown spectrum, as well as the normal “Self-Training Interpretative and Retrieval System” match-factor results, requires 68 s on a laboratory computer. The method is also available to outside users on an international computer network.     

Signatures of the hydrogen bonding in the infrared bands of water

Infrared spectroscopy measurements have been completed over a wide range of frequencies allowing to measure the evolution of both intramolecular and intermolecular vibrational modes in water as a function of temperature. Emphasis is made on the high frequency OH stretching band and the so-called connectivity band that lies in the far infrared region. The substructures of the two infrared bands are analyzed in terms of different levels of connectivity of the water molecules, along the statements of the percolation model. Both band profiles appear to be related to the different degrees of connectivity of water molecules. Comparison of the data with the predictions of the percolation model shows good agreement as for the temperature evolution of liquid water. This work provides additional support to the interpretation of water bands substructures as signatures of its very specific connectivity pattern.     

Study of thermally treated dickite by infrared and micro-Raman spectroscopy using curve-fitting technique

The products of dickite heated in air at 1000 to 1300°C were studied using curve-fitting of transmission and photoacoustic infrared and micro-Raman spectra. The spectra were compared with those of mullite, Al-spinel, corundum, cristobalite, amorphous silica and meta-dickite. Bands that characterize crystalline phases appeared at 1100°C and became stronger with increasing temperature. Mullite, Al-spinel, corundum and amorphous silica were identified by their characteristic bands. The characteristic IR bands of cristobalite overlap those of mullite and amorphous silica, and its presence was therefore established from intensity ratios of the appropriate bands. The research clearly demonstrated the advantage of using curve-fitting for the identification of high temperature phases in the study of the thermal treatment of kaolin-like minerals by infrared and Raman spectroscopy. This technique seems to be a useful method for materials analysis in the ceramic industry.This revised version was published online in November 2005 with corrections to the Cover Date.     

Dynamische interpretation von infrarotspektren

(Dynamic Interpretation of Infrared Spectra) Results of investigation of an algorithm, based on a novel static-dynamic method, for interpretation of infrared spectra are presented. The IRSCAN-D algorithm can recognize 103 infrared-active substructures comprising C, H, N, O, S, halogen, P and Si atoms. The results indicate that the algorithm is efficient; roughly 97% of the substructures are correctly ascribed, with a surprisingly low number (60%) of false recognition.     

Identification of the commonest cystic fibrosis transmembrane regulator gene DeltaF508 mutation: evaluation of PCR--single-strand conformational polymorphism and polyacrylamide gel electrophoresis

In the present study we investigated whether single-strand conformational polymorphism (SSCP) and polyacrylamide gel electrophoresis (PAGE) could be used for the identification of the CFTR DeltaF508 gene mutation, which is commonest in the Greek population. Using DNA from patients carrying this mutation, the appropriate 98 bp region of the CFTR gene was amplified by PCR and the reaction products were analysed by non-radioactive SSCP-electrophoresis using silver staining for band visualization and non-denaturating PAGE to confirm the results. SSCP electrophoretic analysis has been optimized for several parameters in order to achieve the best resolution. Single-strand DNA fragments gave a reproducible pattern of bands, characteristic for the particular mutation. Comparison of the obtain patterns with control samples allowed the detection of the DeltaF508 mutation in the patients studied by SSCP assay and these results were confirmed by the independent method of PAGE. Although SSCP and PAGE can be used for detection of this mutation, PAGE resulted in more distinct patterns than SSCP. It is, therefore, proposed that PAGE can be reliably used for the detection and identification of such a mutation in patients provided that suitable controls are available. The applicability of PAGE to identification of the mutation in carriers, particularly useful for population screening, is also discussed.     

Cyclic voltammetric studies of certain industrially potential iron chelate catalysts

 For the analysis of infrared spectroscopic bands and complex patterns partial cross correlation functions of a sample spectrum with reference spectra are calculated. The chosen ranges of the spectra are based on empirical knowledge of infrared spectrum structure correlations. The normalised maxima of the partial cross correlation functions are interpreted as fuzzy truth values and are combined by fuzzy logical operators. By application of that procedure larger common substructures will be derived from the reference spectra than by a maximum common substructure search based on the complete spectra. Received: 30 October 1996/Revised: 24 February 1997/Accepted: 26 February 1997     

Exhaustive enumeration of molecular substructures

This article addresses the systematic and complete enumeration of all the substructures of any size present in a given molecule. The study is not restricted to features which could be defined a priori such as rings or chains. Contrary to prior expectation the exhaustive enumeration is tractable with current computational tools. Results are presented for several families of skeletons which are widespread in chemistry. It is shown that the numbers of constituent substructures of each size are related to the molecular topology, in particular the degree of branching. The number substructures which are distinct depends additionally on the number of different atom and bond types present. The overall shapes of the distribution of substructure counts as a function of substructure size are found to be similar within particular classes of molecules. These distributions are compared and found to be characteristic of certain topologies. For several simple classes of molecule, analytic expressions are provided for the numbers of substructures as a function of fragment and molecule size. These results hold promise for identifying potentially useful scaffolds for use in combinatorial chemistry. © 1997 by John Wiley & Sons, Inc.     

Preparative thin-layer chromatographic separation and subsequent gas chromatographic-mass spectrometric analysis of monoacylglycerols derived from butter oil by fungal degradation.

A semi-micro method has been developed using preparative thin-layer chromatography (TLC) to separate acylglycerols for the subsequent analysis by gas chromatography-mass spectrometry (GC-MS). Monoacylglycerols (MAGs) were formed from butter oil by fungal degradation with Penicillium roquefortii. Total lipids were extracted with hexane-2-propanol (3:2, v/v) and separated on silica gel preparative TLC plates with fluorescence indicator (Merck). The plates were developed in hexane-diethyl ether-formic acid (80:20:2, v/v). Lipid bands were detected under UV light or with iodine vapour, removed and then extracted with hexane-2-propanol (3:2, v/v). The MAG band (RF 0.03) was silylated into trimethylsilyl (TMS) ethers. Structures and composition of MAG-TMS ethers were analysed by GC and GC-MS. Formation of characteristic ions for the identification of sn-1(3)- and sn-2-MAG isomers was discussed. The method is simple, inexpensive and powerful for the separation and analysis of relatively small amounts of MAGs (0.2-5.0 mg) formed from fungal degradation.     

Combination of the advantages of chromatographic methods based on active components for the quality evaluation of licorice

A rapid, improved and comprehensive method including high‐performance thin‐layer chromatography, fingerprint technology and single standard to determine multiple components was developed and validated for the quality evaluation of licorice. In this study, a newly developed high‐performance thin‐layer chromatography method was first used for authentication of licorice, which achieved simultaneous identification of multiple bands including five bands for known bioactive components by comparing their retention factor values and colors with the standards. For fingerprint analysis, 8 of 16 common peaks were identified. Simultaneously, similarity analysis which showed very similar patterns and hierarchical clustering analysis were performed to discriminate and classify the 27 batches of samples. Additionally, the single standard to determine multiple components method was first successfully achieved to quantify the eight important active markers in licorice including liquiritin apioside, liquiritin, isoliquiritin apioside, isoliquritin, neoisoliquiritin, liquiritigenin, isoliquiritigenin and glycyrrhizic acid. The easily available glycyrrhizic acid was selected as the reference substance to calculate relative response factors. Compared with the normal external standard method, this alternative method can be used to determine the multiple indices effectively and accurately. The validation result showed that the developed method was specific, accurate, precise, robust and reliable for the overall quality assessment of licorice.     

GC-IMS技术结合化学计量学方法在食用植物油分类中的应用

建立了一种快速、无损分析食用植物油中挥发性有机物质的顶空进样/气相色谱-离子迁移谱(GC-IMS)联用方法。以芝麻油、菜籽油、山茶油共56个样品为研究对象,量取2 mL待测油样于标准样品瓶中,并用磁帽密封,直接进行GC-IMS分析检测。结果表明,基于GC-IMS三维谱中对应挥发性有机物质的特征峰强度可以有效表征不同类植物油的样品信息,选取对应三维谱中40个特征峰的强度作为变量,进行主成分(PCA)信息降维后,采用k最近邻(kNN)算法建立植物油种类的判别模型,训练集的识别率达到100%,预测集中仅有1个山茶油样品被误判成芝麻油样品,预测集的识别率达到94.44%。GC-IMS联用分析技术简单、快速、无损,可用于食用植物油等其他食品、农产品种类的快速分类识别。     

Computer-based system for correlating molecular structures with mass spectral data Methods for generating molecular substructures and for incorporating fragmentation rules

The masses of ions observed in the mass spectrum of a pure compound are correlated with the masses of the molecular substructures of the compound. Three methods are described for generating molecular substructures. Each method is evaluated to establish how effectively it generates the molecular substructures and correlates the masses of the molecular substructures with the masses of the observed fragment ions. Rules for mass-spectral fragmentation processes are incorporated into the mass spectral analysis software and illustrated for retro-aldol and lactone-ester reactions occurring in the thermospray mass spectra of oligomycin antibiotics.     

Asbestos mineral analysis by UV Raman and energy-dispersive X-ray spectroscopy.

The applicability of a UV micro-Raman setup was assessed for the rapid identification of fibrous asbestos minerals using 257 and 244 nm laser light for excitation. Raman spectra were obtained from six asbestos reference standards belonging to two basic structural groups: the serpentines (chrysotile) and the amphiboles (crocidolite, tremolite, amosite, anthophyllite, and actinolite). The UV Raman spectra reported here for the first time are free from fluorescence, which is especially helpful in assessing the hydroxyl-stretching vibrations. The spectra exhibit sharp bands characteristic of each asbestos species, which can be used for the unambiguous identification of known and unknown asbestos fibres. Evident changes of the relative band intensities sensitively reflect the chemical substitutions that typically occur in asbestos minerals. The elemental composition of the asbestos reference samples was analysed by using a scanning electron microscope equipped with an energy-dispersive X-ray (EDX) spectrometer. The discussion of the experimental results in terms of EDX analysis sheds new light on the structural and vibrational consequences of cation distribution in asbestos minerals.