A new issue of this journal has just been published. To see abstracts of the papers it contains (with links through to the full papers) click here:
Selected papers from the latest issue:
Identification of primary tumors of brain metastases by Raman Imaging and Support Vector Machines
25 February 2012, 22:07:23
Publication year: 2012
Source: Chemometrics and Intelligent Laboratory Systems, Available online 24 February 2012
Norbert Bergner, Thomas Bocklitz, Bernd F.M. Romeike, Rupert Reichart, Rolf Kalff, ...
Vibrational spectroscopic imaging techniques are new tools for visualizing chemical components in tissue without staining. The spectroscopic signature can be used as a molecular fingerprint of pathological tissues. Fourier transform infrared imaging which is more common than Raman imaging so far has already been applied to identify the primary tumor of brain metastases. The current study introduces a two level discrimination model for Raman microspectroscopic images to distinguish normal brain, necrosis and tumor tissue, and subsequently to determine the primary tumor. 22 specimens of normal brain tissue and brain metastasis of bladder carcinoma, lung carcinoma, mamma carcinoma, colon carcinoma prostate carcinoma and renal cell carcinoma were snap frozen, and thin tissue sections were prepared. Raman microscopic images were collected with 785 nm laser excitation at 10 μm step size. Cluster analysis, vertex component analysis and principal component analysis were applied for data preprocessing. Then, data of 17 specimens were used to train the discrimination model based on support vector machines with radial basis functions kernel. The training data were discriminated with accuracy better than 99%. Finally, the discrimination model correctly predicted independent specimens. The results were superior to discrimination by partial least squares discriminant analysis and support vector machines with linear basis function kernel that were applied for comparison.
Source: Chemometrics and Intelligent Laboratory Systems, Available online 24 February 2012
Norbert Bergner, Thomas Bocklitz, Bernd F.M. Romeike, Rupert Reichart, Rolf Kalff, ...
Vibrational spectroscopic imaging techniques are new tools for visualizing chemical components in tissue without staining. The spectroscopic signature can be used as a molecular fingerprint of pathological tissues. Fourier transform infrared imaging which is more common than Raman imaging so far has already been applied to identify the primary tumor of brain metastases. The current study introduces a two level discrimination model for Raman microspectroscopic images to distinguish normal brain, necrosis and tumor tissue, and subsequently to determine the primary tumor. 22 specimens of normal brain tissue and brain metastasis of bladder carcinoma, lung carcinoma, mamma carcinoma, colon carcinoma prostate carcinoma and renal cell carcinoma were snap frozen, and thin tissue sections were prepared. Raman microscopic images were collected with 785 nm laser excitation at 10 μm step size. Cluster analysis, vertex component analysis and principal component analysis were applied for data preprocessing. Then, data of 17 specimens were used to train the discrimination model based on support vector machines with radial basis functions kernel. The training data were discriminated with accuracy better than 99%. Finally, the discrimination model correctly predicted independent specimens. The results were superior to discrimination by partial least squares discriminant analysis and support vector machines with linear basis function kernel that were applied for comparison.
Kernelk-nearest neighbor algorithm as a flexible SAR modeling tool
25 February 2012, 22:07:23
Publication year: 2012
Source: Chemometrics and Intelligent Laboratory Systems, Available online 24 February 2012
Dong-Sheng Cao, Jian-Hua Huang, Jun Yan, Liang-Xiao Zhang, Qian-Nan Hu, ...
A kernel version ofk-nearest neighbor algorithm (k-NN) has been developed to model the complex relationship between molecular descriptors and bioactivities of compounds. Kernelk-NN is to perform the originalk-NN algorithm by mapping the training samples in the input space into a high-dimensional feature space. It can be easily constructed by calculating the distance between samples in the feature space, directly deriving from the simple calculation of the kernel used. The developed kernelk-NN is very flexible to deal with complex nonlinear relationship, more importantly; it can also conveniently cope with some non-vectorial data only by the definition of different kernels. The results obtained from several real SAR datasets indicated that the performance of kernelk-NN is comparable to support vector machine methods. It can be regarded as an alternative modeling technique for several chemical problems including the study of structure-activity relationship (SAR). The source codes implementing kernelk-NN in R language are freely available athttp://code.google.com/p/kernelmethods/.
Source: Chemometrics and Intelligent Laboratory Systems, Available online 24 February 2012
Dong-Sheng Cao, Jian-Hua Huang, Jun Yan, Liang-Xiao Zhang, Qian-Nan Hu, ...
A kernel version ofk-nearest neighbor algorithm (k-NN) has been developed to model the complex relationship between molecular descriptors and bioactivities of compounds. Kernelk-NN is to perform the originalk-NN algorithm by mapping the training samples in the input space into a high-dimensional feature space. It can be easily constructed by calculating the distance between samples in the feature space, directly deriving from the simple calculation of the kernel used. The developed kernelk-NN is very flexible to deal with complex nonlinear relationship, more importantly; it can also conveniently cope with some non-vectorial data only by the definition of different kernels. The results obtained from several real SAR datasets indicated that the performance of kernelk-NN is comparable to support vector machine methods. It can be regarded as an alternative modeling technique for several chemical problems including the study of structure-activity relationship (SAR). The source codes implementing kernelk-NN in R language are freely available athttp://code.google.com/p/kernelmethods/.
Highlights
► A kernel version of k-NN has been developed. ► The performance of kernel k-NN is competitive to one by SVM. ► Kernel k-NN can cope with non-vectorial data such as string data etc. ► Weighted kernel k-NN was developed to allow the construction of ROC.Extracting homologous series from mass spectrometry data by projection on predefined vectors
25 February 2012, 22:07:23
Publication year: 2012
Source: Chemometrics and Intelligent Laboratory Systems, Available online 23 February 2012
Johan E. Carlson, James R. Gasson, Tanja Barth, Ingvar Eide
Multivariate statistical methods, such as Principal Component Analysis (PCA), have been used extensively over the past decades as tools for extracting significant information from complex data sets. As such they are very powerful and in combination with an understanding of underlying chemical principles, they have enabled researchers to develop useful models. A drawback with the methods is that they do not have the ability to incorporate any physical / chemical model of the system being studied during the statistical analysis. In this paper we present a method that can be used as a complement to traditional chemometric tools in finding patterns in mass spectrometry data. The method uses a pre-defined set of equally spaced sequences that are assumed to be present in the data. Allowing for some uncertainty in the peak locations due to the uncertainties for the measurement instrumentation, the measured spectra are then projected onto this set. It is shown that the resulting scores can be used to identify homologous series in measured mass spectra that differ significantly between different measured samples. As opposed to PCA, the loading vectors, in this case the pre-defined homologous series, are readily interpretable.
Source: Chemometrics and Intelligent Laboratory Systems, Available online 23 February 2012
Johan E. Carlson, James R. Gasson, Tanja Barth, Ingvar Eide
Multivariate statistical methods, such as Principal Component Analysis (PCA), have been used extensively over the past decades as tools for extracting significant information from complex data sets. As such they are very powerful and in combination with an understanding of underlying chemical principles, they have enabled researchers to develop useful models. A drawback with the methods is that they do not have the ability to incorporate any physical / chemical model of the system being studied during the statistical analysis. In this paper we present a method that can be used as a complement to traditional chemometric tools in finding patterns in mass spectrometry data. The method uses a pre-defined set of equally spaced sequences that are assumed to be present in the data. Allowing for some uncertainty in the peak locations due to the uncertainties for the measurement instrumentation, the measured spectra are then projected onto this set. It is shown that the resulting scores can be used to identify homologous series in measured mass spectra that differ significantly between different measured samples. As opposed to PCA, the loading vectors, in this case the pre-defined homologous series, are readily interpretable.
No comments:
Post a Comment