Identification of peptides within a known protein sequence using COMSEQ analysis of data containing multiple sequences

Modern methods of automated protein sequence analysis can provide high-quality data with which unambiguous amino-acid sequences can be determined, but analyses are more difficult when the sample is not pure. COMSEQ and auxillary programs were written to facilitate reconciliation of multiple amino-acid sequences potentially contained in noisy data with the known amino-acid sequence of the parent protein. The COMSEQ program prints a matrix in which the first vertical column represents the known amino-acid sequence of a selected protein. Each row of the matrix contains the sequencer yield corresponding to the amino acid in the first column, with each column corresponding to the sequencing reaction cycle. A diagonal which contains net increases of amino acids for each amino acid in the known sequence identifies a peptide potentially contained within the data. The number of matches for each diagonal over the entire known sequence are tabulated and presented as an aid to locating comparisons of greatest interest. The RNDSEQ program conducts multiple analyses using randomized versions of the known amino-acid sequence and tabulates the cumulative frequencies of potential sequence matches irrespective of the true known sequence. TRANSEQ is a utility program that translates edited sequence data from common databases into files that can be used by COMSEQ and RNDSEQ. The programs have been used successfully to identify two co-sequenced peptides from bovine serum albumin, an albumin peptide sequence in the presence of hemoglobin, and to identify two sequences of rat alpha-2u-globulin that differ in their amino termini.     

Purification and lipid dependence of the recombinant hyaluronan synthases from Streptococcus pyogenes and Streptococcus equisimilis

The two hyaluronan synthases (HASs) from Streptococcus pyogenes (spHAS) and Streptococcus equisimilis (seHAS) were expressed in Escherichia coli as recombinant proteins containing His6 tails. Both enzymes were expressed as major membrane proteins, accounting for approximately 5-8% of the total membrane protein. Using nickel chelate affinity chromatography, the HASs were purified to homogeneity from n-dodecyl beta-D-maltoside extracts. High levels of HAS activity could be achieved only if the purified enzymes were supplemented with either bovine or E. coli cardiolipin (CL), although bovine CL gave consistently greater activity. Mass spectroscopic analysis revealed that the fatty acid compositions of these two CL preparations did not overlap. The two HAS enzymes showed similar but distinct activation profiles with the 10 other lipids tested. For example, phosphatidic acid and phosphatidylethanolamine stimulated seHAS, but not spHAS. Phosphatidylserine stimulated both enzymes. spHAS appears to be more CL-specific than seHAS, although both purified enzymes still contain endogenous CL that can not easily be removed. Both seHAS and spHAS were inhibited by phosphatidylcholine, sphingomyelin, and sulfatides and were not substantially stimulated by cerebrosides, phosphatidylglycerol, or phosphatidylinositol. With both HASs, CL increased the Km for UDP-GlcUA, but decreased the Km for UDP-GlcNAc and gave an overall stimulation of Vmax. A kinetic characterization of the two membrane-bound and purified HASs is presented in the accompanying paper (Tlapak-Simmons, V. L., Baggenstoss, B. A., Kumari, K., Heldermon, C., and Weigel, P. H. (1999) J. Biol. Chem. 274, 4246-4253). Both purified HASs became inactive after storage for approximately 5 days at 4 degreesC. Both purified enzymes also lost activity over 4-5 days when stored at -80 degreesC in the presence of CL, but reached a level of activity that then slowly decreased over a period of months. Although the purified enzymes stored in the absence of CL at -80 degreesC were much less active, the enzymes retained this same low level of activity for at least 5 weeks. When both spHAS and seHAS were stored without CL at -80 degreesC, even after 2 months, they could be stimulated by the addition of bovine CL to approximately 60% of the initial activity of the freshly purified enzyme.     

Multidimensional probability density function approximations fordetection, classification, and model order selection

This paper addresses the problem of calculating the multidimensional probability density functions (PDFs) of statistics derived from known many-to-one transformations of independent random variables (RVs) with known distributions. The statistics covered in the paper include reflection coefficients, autocorrelation estimates, cepstral coefficients, and general linear functions of independent RVs. Through PDF transformation, these results Can be used for general PDF approximation, detection, classification, and model order selection. A model order selection example that shows significantly better performance than the Akaike and MDL method is included     

The PDF projection theorem and the class-specific method

We present the theoretical foundation for optimal classification using class-specific features and provide examples of its use. A new probability density function (PDF) projection theorem makes it possible to project probability density functions from a low-dimensional feature space back to the raw data space. An M-ary classifier is constructed by estimating the PDFs of class-specific features, then transforming each PDF back to the raw data space where they can be fairly compared. Although statistical sufficiency is not a requirement, the classifier thus constructed becomes equivalent to the optimal Bayes classifier if the features meet sufficiency requirements individually for each class. This classifier is completely modular and avoids the dimensionality curse associated with large complex problems. By recursive application of the projection theorem, it is possible to analyze complex signal processing chains. We apply the method to feature sets, including linear functions of independent random variables, cepstrum, and Mel cepstrum. In addition, we demonstrate how it is possible to automate the feature and model selection process by direct comparison of log-likelihood values on the common raw data domain.     

An adaptive detector for deterministic signals in noise of unknownspectra using the Rao test

An adaptive detector for a known deterministic signal of unknown amplitude in Gaussian noise of unknown spectra is described. The detector is based on the Rao test, which is asymptotically equivalent to the generalized likelihood ratio. The detector achieves constant false alarm probability in the presence of large changes in input noise bandwidth and variance while providing optimum detection performance. The results are supported by simulation     

Aroma recovery from roasted coffee by wet grinding

Aroma recovery as determined by solid phase microextraction-gas chromatography-mass spectrometry (SPME-GC-MS) was compared in coffees resulting from conventional grinding processes, and from wet grinding with cold and hot water. Freshly roasted coffee as well as old, completely degassed coffee was ground in order to estimate the relationship of internal carbon dioxide pressure in freshly roasted coffee with the aroma loss during grinding. The release of volatile aroma substances during grinding was found to be related to the internal carbon dioxide pressure, and wet grinding with cold water was shown to minimize losses of aroma compounds by trapping them in water. Due to the high solubility of roasted coffee in water, the use of wet-grinding equipment is limited to processes where grinding is followed by an extraction step. Combining grinding and extraction by the use of hot water for wet grinding resulted in considerable losses of aroma compounds because of the prolonged heat impact. Therefore, a more promising two-step process involving cold wet grinding and subsequent hot extraction in a closed system was introduced. The yield of aroma compounds in the resulting coffee was substantially higher compared to conventionally ground coffee.     

Kinetic characterization of the recombinant hyaluronan synthases from Streptococcus pyogenes and Streptococcus equisimilis

The two hyaluronan synthases (HASs) from Streptococcus pyogenes (spHAS) and Streptococcus equisimilis (seHAS) were expressed in Escherichia coli as recombinant proteins containing His6 tails. The accompanying paper has described the purification and lipid dependence of both HASs, their preference for cardiolipin, and their stability during storage (Tlapak-Simmons, V. L., Baggenstoss, B. A., Clyne, T., and Weigel, P. H. (1999) J. Biol. Chem. 274, 4239-4245). Kinetic characterization of the enzymes in isolated membranes gave Km values for UDP-GlcUA of 40 +/- 4 microM for spHAS and 51 +/- 5 microM for seHAS. In both cases, the Vmax profiles at various concentrations of UDP-GlcNAc were hyperbolic, with no evidence of cooperativity. In contrast, membrane-bound spHAS, but not seHAS, showed sigmoidal behavior as the UDP-GlcNAc concentration was increased, with a Hill number of approximately 2, indicating significant cooperativity. The Hill number for UDP-GlcNAc utilization by seHAS was 1, confirming the lack of cooperativity for UDP-GlcNAc in this enzyme. The Km values for UDP-GlcNAc were 60 +/- 7 microM for seHAS and 149 +/- 3 microM for spHAS in the isolated membranes. The kinetic characteristics of the two affinity-purified HAS enzymes were assessed in the presence of cardiolipin after 8-9 days of storage at -80 degreesC without cardiolipin. With increasing storage time, the enzymes showed a gradual increase in their Km values for both substrates and a decrease in Vmax. Even in the presence of cardiolipin, the detergent-solubilized, purified HASs had substantially higher Km values for both substrates than the membrane-bound enzymes. The KUDP-GlcUA for purified spHAS and seHAS increased 2-4-fold. The KUDP-GlcNAc for spHAS and seHAS increased 4- and 5-fold, respectively. Despite the higher Km values, the Vmax values for the purified HASs were only approximately 50% lower than those for the membrane-bound enzymes. Significantly, purified spHAS displayed the same cooperative interaction with UDP-GlcNAc (nH approximately 2), whereas purified seHAS showed no cooperativity.     

Image distortion analysis using polynomial series expansion

In this paper, we derive a technique for analysis of local distortions which affect data in real-world applications. In the paper, we focus on image data, specifically handwritten characters. Given a reference image and a distorted copy of it, the method is able to efficiently determine the rotations, translations, scaling, and any other distortions that have been applied. Because the method is robust, it is also able to estimate distortions for two unrelated images, thus determining the distortions that would be required to cause the two images to resemble each other. The approach is based on a polynomial series expansion using matrix powers of linear transformation matrices. The technique has applications in pattern recognition in the presence of distortions.     

Water content of roasted coffee: impact on grinding behaviour,extraction, and aroma retention

Normal and long time roasting trials were carried out on industrial scale. Different amounts of water were applied during quenching, resulting in water contents in the range of 2.3–8.8 g/100 g wb. Coffees were ground immediately after cooling, and after equilibration times of 6 and 24 h. Particle size distribution of ground coffees, percolation time, and extraction properties were investigated on an espresso coffee machine. Coffees ground after 24 h resting time were subjected to storage trials to determine aroma stability as influenced by water content. Coffees with high moisture content exhibited coarser particles upon grinding, and equilibration time prior to grinding was needed for coffees with high water content to improve grinding results. Coffees with low water content did not exhibit this time dependency prior to grinding. Coffees with low water content were extracted more effectively than high moisture coffees, and percolation was slower. During open and closed storage, evolution of hexanal and sulfides was highly sensitive to water content. However, differences in evolution of other aroma compounds were found during closed storage only, where moisture content had a negative impact on aroma stability of the coffees subjected to investigation.     

On estimating the angle parameters of an exponential signal at highSNR

It is shown that at high signal-to-noise ratio, phase data alone constitute a sufficient statistic for the estimation of the angle parameters of a single exponential signal in Gaussian noise. The threshold effect, which occurs when estimating frequency based on phase only, can be deduced by examining the joint probability distribution function of the phase and amplitude