用四种离散变换主组分回归方法分辨重叠吸收光谱

将离散小波变换、小波包变换、傅里叶变换和离散余弦变换与主组分回归方法结合构成4种离散变换主组分回归方法,编制了离散变换主组分回归方法的计算程序。将离散变换主组分回归方法用于处理对硝基甲苯、对硝基酚和对硝基苯胺混合物的重叠紫外吸收光谱数据。结果表明,离散变换主组分回归方法优于主组分回归方法,试样质量浓度的预测值与实际值的相对预测标准误差由3.81%降至约1.11%。     

一类用于多元光度分析的小波基主成分回归法

通过将变尺度小波分解滤噪与特征信息提取相结合,提出一类新的多元光度分析算法－小波基主成分回归（ＰＣＲＷ）方法。该法可有效地减小主成分向量残留噪声所引 误差显著提高多元校正准确性。将其用于分析氯霉素、醋酸对塞米松、尼泊金乙脂体系、得到满意的回收率,与ＰＣＲ法相比,分析结果的总平均相对误差从３．３８％降低到０．８３％。     

Daubechies小波主成分回归法机理及算法研究

将小波变换与主成分回归相结合,提出一种新型多元校正算法---小波基主成分回归法。理论分析和仿真实验表明,该法可更有效地去除噪声,提取有用信息。将其用于氯霉素及甲硝唑实际药物体系分析,与主成分回归法(PCR)相比,得到的回收率总平均相对误差由1.70%下降到0.90%。此外,通过将统计判据和小波多尺度分析相结合,发展了一种新的因子数判定方法。理论和实验研究表明,该法比传统因子数判定法具有更高的可靠性。     

基于小波变换平滑主成分分析

小波变换具有很强的信号分离能力,很容易把随机噪音从信号中分离出来,从而提高信号的信噪比。本文把小波变换引入到因子分析中,提出了基于小波变换平滑主成分分析,该算法既保留普通主成分分析的正交分解,又具备了小波变换的信号分离能力。模拟数据和实验数据的结果表明,该算法具有从低信噪比的数据中提取出有用信息,并提高信号的信噪比。迭代目标变换因子分析处理实验数据的结果表明,基于小波变换平滑主成分分析的处理结果优     

连续小波变换-独立成分回归算法及其在多组分分析中的应用

采用连续小波变换（CWT）对光谱数据进行处理，用独立成分分析（ICA）进行特征提取，再用回归分析方法对被测组分进行测定，建立了连续小波变换一独立成分回归（CWT-ICR）方法。方法用于肉样品中水分、脂肪和蛋白质多组分的同时测定，所得结果与化学法测得结果相符。     

基于料酒的三维同步荧光光谱模式识别研究

提出了一种应用同步荧光光谱技术无损快速鉴别料酒品牌的新方法.利用主成分分解法和小波变换法对料酒样品的同步荧光光谱信号进行了压缩处理,分别采用同步荧光光谱数据的第一主成分和小波细节系数为特征变量进行主成分分析和聚类分析,分类结果表明小波系数作为料酒的特征变量对料酒品牌分类正确率更高.利用偏最小二乘和径向基人工神经网络方法...     

小波包变换潜变量回归分辨重叠的紫外光谱

采用小波包变换潜变量回归(WPLVR)方法,同时测定联苯、苯酚和邻苯二酚。该法结合小波包变换和潜变量回归改进除噪质量。通过最佳化,选择了小波函数及小波包分解水平(L)。编制了两个程序PWPLVR和PFTLVR进行WPLVR和付立叶变换潜变量回归(FTLVR)法计算。试验结果表明WPLVR法是成功的且优于FTLVR法。     

同时测定合成样中铜钴镍钒含量的分光光度法--主成分回归法和偏最小二乘法比较

研究了主成分回归和偏最小二乘方法在多组分分光光度法分析中的应用,以５－Ｂｒ－ＰＡＤＡＰ（２〔５－溴－２－吡啶）－偶氮〕５－二乙氨基苯酚）为显色剂,ＯＰ（聚乙二醇辛基苯基醚）作为增溶增稳剂,在ｐＨ＝３．６的条件下,用主成分回归及偏最小二乘分光光度法同时测定了合成样中的铜、钴、镍、钒４组分含量,测定相对误差在－６．００％～４．００％之间。实验证明,对于加和性不好的体系偏最小二乘分光光度法要优于主成分回     

小波包变换潜变量回归同时测定三组分混合物

采用小波包变换潜变量回归 (WPLVR)方法 ,同时测定水杨酸甲酯(MSA)、邻苯二甲酸二丁酯 (DBP)和邻苯二甲酸氢钾 (PHP)。该法结合小波包变换和潜变量回归改进除噪质量。通过最佳化 ,选择了小波函数及小波包分解水平 (L)。编制了两个程序 (PWPLVR)和 (PFTLVR)进行WPLVR和付立叶变换潜变量回归 (FTLVR)法计算。实验结果表明WPLVR法是成功的且优于FTLVR法。     

近红外漫反射光谱的小波变换滤波

利用小波变换对５２个烟草样品的近红外漫反射光谱进行滤波处理，并用ＰＬＳ法来计算烟草样品的总氮含量，结果表明小波变换滤波后，预测集的相对标准偏差由原来的９．２％降为７．４％，此结果也优于傅里叶变换和五点三次平滑。     

Analysis of electrophoretic patterns of arbitrarily primed PCR profiling

We present a mathematical algorithm for the analysis of electrophoretic patterns resulting from arbitrarily primed PCR profiling. The algorithm is based on the established mathematical procedures applied to the analysis of digital images of gel patterns. The algorithm includes (a) transformation of the image into a matrix form, (b) identification of every electrophoretic lane as a set of matrix columns that are further mathematically processed, (c) averaging of matrix columns corresponding to electrophoretic lanes that define lane representatives, (d) elimination of "smiling" bands, (e) solving the problem of a lane offset, and (f) removal of the background. Representation of individual electrophoretic lanes in the form of functions allows interlane comparisons and further mathematical analysis. Direct comparison of selected lanes was obtained by employing correlation analysis. Gel images were those obtained after arbitrarily primed PCR analysis of DNA that underwent damage induced by gamma radiation from a (60)Co source. The applied method proved to be useful for elimination of subjectivity of visual inspection. It offers the possibility to avoid overlooking important differences in case of suboptimal electrophoretic resolution. In addition, higher precision is achieved in the assessment of quantitative differences due to better insight into experimental artifacts. These simple mathematical methods offer an open-type algorithm, i.e., this algorithm enables easy implementation of different parameters that may be useful for other analytical needs.     

近红外光谱法测定端羟基环氧乙烷-四氢呋喃共聚醚的羟值

提出了用近红外光谱测定端羟基环氧乙烷-四氢呋喃共聚醚(PET)的羟值,结合主成分回归和偏最小二乘法建立了PET羟值与其近红外光谱之间的关联模型。结果表明,近红外光谱法与化学分析法的测定结果一致;近红外光谱法测定PET羟值的相对误差在5%以内;利用遗传算法选择部分波长建立校正可以降低模型的预测误差。     

Variable predictive model based classification algorithm for effective separation of protein structural classes

Variable predictive model based class discrimination (VPMCD) algorithm is proposed as an effective protein secondary structure classification tool. The algorithm mathematically represents the characteristics amino acid interactions specific to each protein structure and exploits them further to distinguish different structures. The new concept and the VPMCD classifier are established using well-studied datasets containing four protein classes as benchmark. The protein samples selected from SCOP and PDB databases with varying homology (25-100%) and non-uniform distribution of class samples provide challenging classification problem. The performance of the new method is compared with advanced classification algorithms like component coupled, SVM and neural networks. VPMCD provides superior performance for high homology datasets. 100% classification is achieved for self-consistency test and an improvement of 5% prediction accuracy is obtained during Jackknife test. The sensitivity of the new algorithm is investigated by varying model structures/types and sequence homology. Simpler to implement VPMCD algorithm is observed to be a robust classification technique and shows potential for effective extensions to other clinical diagnosis and data mining applications in biological systems.     

A new algorithm for the determination of protolytic constants from spectrophotometric data in multiwavelength mode: Calculations of acidity constants of 4-(2-pyridylazo)resorcinol (PAR) in mixed nonaqueous-water solvents

A new efficient, simple and versatile algorithm is presented for determination of the protolytic constants from spectrophotmetric data in multiwavelength mode based on the combining of hard and soft modeling. The algorithm was checked by determining the acidity constants of a triprotic acid from theoretical and real absorption-pH data. The real spectral data are obtained from photometric titration of different solutions of 4-(2-pyridylazo)resorcinol (PAR) by a standard base solution under an inert atmosphere. The algorithm starts the minimization process using an user supplied number of components and initial guesses of the unknown parameters and refined in a least squares manner. New algorithm is implemented in the new version of DATAN package (version 3.1). The validity of the obtained results was checked by some well known programs such as DATAN 2.1, SPECFIT/32, SQUAD, a modified version of difference spectra and a A-pH curve method. The comparison of the outputs of the DATAN 3.1 with the other programs reveals that there is a very good agreement between the obtained results and mentioned programs.     

Simultaneous determination of composition of human urinary calculi by use of artificial neural networks

A new chemometric method, which uses artificial neural networks (ANN), is presented for determination of the composition of urinary calculi. The selected constituents were whewellite, weddellite, and uric acid from which approximately 40% of the urinary calculi obtained from Macedonia patients are composed. The results for the synthetic mixtures were better then those obtained by partial least squares (PLS) regression or by the principal component regression (PCR), because neural networks have better prediction capacity. The generalization abilities of the optimized neural networks were checked using the standard addition method on carefully selected real natural samples.     

Byte structure variable length coding (BS-VLC): a new specific algorithm applied in the compression of trajectories generated by molecular dynamics

Molecular dynamics is a well-known technique very much used in the study of biomolecular systems. The trajectory files produced by molecular dynamics simulations are extensive, and the classical lossless algorithms give poor efficiencies in their compression. In this work, a new specific algorithm, named byte structure variable length coding (BS-VLC), is introduced. Trajectory files, obtained by molecular dynamics applied to trypsin and a trypsin:pancreatic trypsin inhibitor complex, were compressed using four classical lossless algorithms (Huffman, adaptive Huffman, LZW, and LZ77) as well as the BS-VLC algorithm. The results obtained show that BS-VLC nearly triplicates the compression efficiency of the best classical lossless algorithm, preserving a near lossless behavior. Compression efficiencies close to 50% can be obtained with a high degree of precision, and the maximum efficiency possible (75%), within this algorithm, can be performed with good precision.     

间接电合成甲基苯甲醛产物中3种异构体的同时测定

将迭代算法、PLS与紫外吸收光谱法相结合,建立了同时测定二甲苯间接电合成甲基苯甲醛混合6组分中3种甲基苯甲醛含量的新方法。该方法将PLS镶嵌在迭代算法内部,通过多次迭代计算逐步逼近样品真值,提高了光谱的识别能力。将该方法用于4组模拟样本的测定,邻、间、对甲基苯甲醛的平均回收率分别为101%、100%和101%,预测均方差(RMSEP)分别为0.214、0.148和0.057。而应用于5批实际电合成产物中3种甲基苯甲醛同分异构体的同时测定,邻、间、对甲基苯甲醛的平均回收率分别为106%、92%和100%,相对偏差≤±19.7%,与高效液相色谱法测定结果基本吻合。     

Evaluation of a polymerase chain reaction-based heteroduplex assay for detecting the adulteration of processed orange juice with mandarin juice

A polymerase chain reaction (PCR)-based heteroduplex assay was evaluated for the detection of mandarin juice in processed orange juice. PCR amplification of a fragment of the chloroplast trnT-trnL intergenic spacer derived from mixtures of DNA extracted from orange and mandarin juice resulted in heteroduplex formation. The heteroduplex resulted from the co-amplification of a fragment containing an 8 base-pair indel that distinguished mixtures of orange and mandarin juice from orange juice and mandarin juice alone. The heteroduplex assay was evaluated against authentic juices obtained from different citrus species and confirmed that the marker was homogeneous within Citrus. The data obtained demonstrated maternal inheritance of chloroplast type in Citrus sp. and allowed the identification and confirmation of the maternal parentage of unknown and known citrus hybrids. Analysis of the quantitative potential of the PCR and polyacrylamide gel electrophoresis (PAGE) analysis demonstrated good repeatability with a coefficient of variation of 7.5%. Greatest sources of variance in experimental results were attributable to species and varietal differences in the levels of the PCR target. Mandarin juice contained approximately 18% (w/v) less PCR target sequence than did orange juice. The assay was tested in a blind trial using processed juices and correctly identified 20/22 samples with no false-positive results.