Prognostic predictor with multiple fuzzy neural models using expression profiles from DNA microarray for metastases of breast cancer

Gene expression profiling data from DNA microarray were analyzed using the fuzzy neural network (FNN) modeling method for predicting the distant metastases of breast cancer. The best model consisting of five genes was able to predict metastases of breast cancer with 94% accuracy. Furthermore, 100% accuracy was achieved by majoritarian decision using only 25 genes from five noninferior models which were constructed independently. From the constructed model, gene expression rules, which may cause distant metastases, were explicitly extracted and 60% of the metastases cases could be explained by this rule. The FNN modeling method described in this paper enables precise extraction of significant biological markers affecting prognosis without prior knowledge.     

Prediction of peptide binding to major histocompatibility complex class II molecules through use of boosted fuzzy classifier with SWEEP operator method

To treat autoimmune diseases, it is important to identify which peptides bind to major histocompatibility complex (MHC) class II molecules (HLA-DRs). Predicting the peptides that bind to MHC class II molecules can effectively reduce the number of experiments required for identifying helper T cell epitopes. In our previous study, we applied fuzzy neural networks (FNNs) to solve this problem. However, an FNN requires a long calculation time and a large number of peptides; this means performing several experiments. In this study, we applied a boosted fuzzy classifier with the SWEEP operator method (BFCS) to solve this problem. For comparison, two other conventional modeling methods, namely, support vector machine and FNN combined with the SWEEP operator method (FNN-SWEEP) instead of using solely an FNN, were employed. Compared with FNN, FNN-SWEEP is extremely fast and has an almost identical prediction accuracy. The model constructed by BFCS showed an accuracy approximately 5%-10% higher than that constructed by FNN-SWEEP. In addition, BFCS was 30,000-120,000 times faster than FNN-SWEEP. This result suggests that BFCS has the potential to function as a new method of predicting peptides that bind to various protein receptors.     

Sensory Modeling of Coffee with a Fuzzy Neural Network

ABSTRACT: Models were constructed to predict sensory evaluation scores from the blending ratio of coffee beans. Twenty-two blended coffees were prepared from 3 representative beans and were evaluated with respect to 10 sensory attributes by 5 coffee cup-tasters and by models constructed using the response surface method (RSM), multiple regression analysis (MRA), and a fuzzy neural network (FNN). The RSM and MRA models showed good correlations for some sensory attributes, but lacked sufficient overall accuracy. The FNN model exhibited high correlations for all attributes, clearly demonstrated the relationships between blending ratio and flavor characteristics, and was accurate enough for practical use. FNN, thus, constitutes a powerful tool for accelerating product development.     

Construction of COD simulation model for activated sludge process by fuzzy neural network

Fuzzy neural network (FNN) was applied to construct a simulation model for estimating the effluent chemical oxygen demand (COD) value of an activated sludge process in a "U" plant, in which most of process variables were measured once an hour. The constructed FNN model could simulate periodic changes in COD with high accuracy. Comparing the simulation result obtained using the FNN model with that obtained using the multiple regression analysis (MRA) model, it was found that the FNN model had 3.7 times higher accuracy than the MRA model. The FNN models corresponding to each of the four seasons were also constructed. Analyzing the fuzzy rules acquired from the FNN models after learning, the operational characteristic of this plant could be elucidated. Construction of the simulation model for another plant "A", in which process variables were measured once a day, was also carried out. This FNN model also had a relatively high accuracy.     

Fuzzy neural network-based prediction of the motif for MHC class II binding peptides

Characterizing the interaction between major histocompatibility complex (MHC) molecules and antigenic peptides is critical for understanding immunity and developing immunotherapies for autoimmune diseases and cancer. To identify the peptide binding motif and predict peptides that bind to the human MHC classII molecule HLA-DR4(*0401), we applied a fuzzy neural network (FNN) capable of extracting the relationship between input and output. Analysis of the peptide binding motif revealed that the hydrophilicity of the position 1 residue located on the N-terminal side of the nonamer (9mer) was the most important variable and that the van der Waals volume and hydrophilicity of the position 6 residue and the hydrophilicity of the position 7 residue were also important variables. The estimation accuracy (A(ROC) value) was high and the binding motif extracted from the FNN agreed with that derived experimentally. This study demonstrates that FNN modeling allows candidate antigenic peptides to be selected without the need for further experiments.     

Comparison of modelling techniques for milk-production forecasting

The objective of this study was to assess the suitability of 3 different modeling techniques for the prediction of total daily herd milk yield from a herd of 140 lactating pasture-based dairy cows over varying forecast horizons. A nonlinear auto-regressive model with exogenous input, a static artificial neural network, and a multiple linear regression model were developed using 3 yr of historical milk-production data. The models predicted the total daily herd milk yield over a full season using a 305-d forecast horizon and 50-, 30-, and 10-d moving piecewise horizons to test the accuracy of the models over long- and short-term periods. All 3 models predicted the daily production levels for a full lactation of 305 d with a percentage root mean square error (RMSE) of ≤12.03%. However, the nonlinear auto-regressive model with exogenous input was capable of increasing its prediction accuracy as the horizon was shortened from 305 to 50, 30, and 10 d [RMSE (%) = 8.59, 8.1, 6.77, 5.84], whereas the static artificial neural network [RMSE (%) = 12.03, 12.15, 11.74, 10.7] and the multiple linear regression model [RMSE (%) = 10.62, 10.68, 10.62, 10.54] were not able to reduce their forecast error over the same horizons to the same extent. For this particular application the nonlinear auto-regressive model with exogenous input can be presented as a more accurate alternative to conventional regression modeling techniques, especially for short-term milk-yield predictions.     

Artificial neural network and mathematical modeling comparative analysis of nonisothermal diffusion of moisture in wood

The objective of this study was to develop an optimum artificial neural network (ANN) capable of predicting the direction and magnitude of the moisture flux through wood under nonisothermal steady-state diffusion. A comparison between experimental measurements and the predicted values of three mathematical models reported in the literature and of the proposed neural network is presented and discussed. When developing the ANN model, several configurations were evaluated. The optimal ANN model was found to be a network with six neurons in one hidden layer. This well-trained network correlated the forecasted to the experimental data with low-level errors compared to previously developed models and also predicted the flux-reversal phenomenon thus confirming that ANN modeling has a much better predictive performance. It was also shown that the numbers of the training data were linked to the performance of the network during estimation. However, the powerful predictive capacity of this modeling method was still supported although a limited experimental data set was trained.     

烟用爆珠内液质量稳定性检测——基于紫外光谱技术结合SVR算法

  目的  基于紫外（UV）光谱技术和支持向量机回归（SVR）算法,建立一种烟用爆珠内液判别和稳定性分析方法。  方法  采用无水乙醇稀释爆珠内液,UV扫描,光谱经预处理后,建立四种不同牌号烟用爆珠SVR模型,并对模型进行了验证。  结果  （1）最佳UV图谱预处理方法是平滑后归一化;最佳SVR类型是ν-SVR;最佳核函数是径向基核函数。（2）四个牌号烟用爆珠内液ν-SVR模型校正集分类变量的预测值与实测值的相关系数均≥0.9993,SVR模型对于参与建模的400个烟用爆珠内液样本（校正集）和未参与建模的80个烟用爆珠内液样本（验证集）的预测准确度均为100%,模型拟合性好,预测精度高,判别能力强。（3）基于SVR模型分类变量值建立的单值控制图可以对烟用爆珠内液的稳定性进行快速判定,判定结果与GC/MS检测结果一致。  结论  紫外可见光谱技术结合SVR算法可对不同牌号烟用爆珠内液质量进行有效判别,且方法快速准确、经济环保、易于推广。      

超高压和热杀菌的蓝莓果汁饮料贮藏期品质的变化及货架期预测模型

比较了热杀菌（80 ℃,10 min）和超高压杀菌（550 MPa,10 min）的蓝莓果汁饮料产品贮藏期（54 d）品质的变化,并且预测了蓝莓果汁饮料的货架期。结果表明:蓝莓果汁饮料的贮藏期结束时,两种杀菌条件的蓝莓果汁饮料在不同贮藏温度（4、27和37 ℃）均未检测出微生物,说明杀菌彻底。蓝莓果汁饮料的pH和可溶性固形物在贮藏期间变化不大。超高压杀菌的蓝莓果汁饮料的品质较好,贮藏结束时蓝莓果汁饮料的抗坏血酸含量和总酚含量都较高。基于蓝莓果汁饮料的感官评分的变化采用动力学模型结合Arrhenius方程建立了4~37 ℃范围内的品质劣变动力学模型及货架期预测模型,并对其预测精确度进行了评价。所建立的模型的决定系数R2均在0.95以上,货架期预测相对误差大多在10%之内。因此,所建立的模型能够快速可靠地预测蓝莓果汁饮料的剩余货架期。     

Comparative quantitative prevalence of Mycobacteria and functionally abundant nidA, nahAc, and nagAc dioxygenase genes in coal tar contaminated sediments

The Chattanooga Creek Superfund site is heavily contaminated with metals, pesticides, and coal tar with sediments exhibiting high concentrations of polycyclic aromatic hydrocarbons (PAHs). High molecular weight PAHs are of concern because of their toxicity and recalcitrance in the environment; as such, there is great interest in microbes, such as fast-growing Mycobacterium spp., capable of degradation of these compounds. Real-time quantitative PCR assays were developed targeting multiple dioxygenase genes to assess the ecology and functional diversity of PAH-degrading communities. These assays target the Mycobacterium nidA, beta-proteobacteria nagAc, and gamma-proteobacteria nahAc with the specific goal of testing the hypothesis that Mycobacteria catabolic genes are enriched and may be functionally associated with high molecular weight PAH biodegradation in Chattanooga Creek. Dioxygenase gene abundances were quantitatively compared to naphthalene and pyrene mineralization, and temporal and spatial PAH concentrations. nidA abundances ranged from 5.69 x 10(4) to 4.92 x 10(6) copies per gram sediment; nagAc from 2.42 x 10(3) to 1.21 x 10(7), and nahAc from below detection to 4.01 x 10(6) copies per gram sediment. There was a significantly greater abundance of nidA and nagAc at sites with the greatest concentrations of PAHs. In addition, nidA and nagAc were significantly positively correlated (r = 0.76), indicating a coexistence of organisms carrying these genes. A positive relationship was also observed between nidA and nagAc and pyrene mineralization indicating that these genes serve as biomarkers for pyrene degradation. A 16S rDNA clone library of fast-growing Mycobacteria indicated that the population is very diverse and likely plays an important role in attenuation of high molecular weight PAHs from Chattanooga Creek.     

Evaluation of incremental reactivity and its uncertainty in Southern California

The incremental reactivity (IR) and relative incremental reactivity (RIR) of carbon monoxide and 30 individual volatile organic compounds (VOC) were estimated for the South Coast Air Basin using two photochemical air quality models: a 3-D, grid-based model and a vertically resolved trajectory model. Both models include an extended version of the SAPRC99 chemical mechanism. For the 3-D modeling, the decoupled direct method (DDM-3D) was used to assess reactivities. The trajectory model was applied to estimate uncertainties in reactivities due to uncertainties in chemical rate parameters, deposition parameters, and emission rates using Monte Carlo analysis with Latin hypercube sampling. For most VOC, RIRs were found to be consistent in rankings with those produced by Carter using a box model. However, 3-D simulations show that coastal regions, upwind of most of the emissions, have comparatively low IR but higher RIR than predicted by box models for C4-C5 alkenes and carbonyls that initiate the production of HOx radicals. Biogenic VOC emissions were found to have a lower RIR than predicted by box model estimates, because emissions of these VOC were mostly downwind of the areas of primary ozone production. Uncertainties in RIR of individual VOC were found to be dominated by uncertainties in the rate parameters of their primary oxidation reactions. The coefficient of variation (COV) of most RIR values ranged from 20% to 30%, whereas the COV of absolute incremental reactivity ranged from about 30% to 40%. In general, uncertainty and variability both decreased when relative rather than absolute reactivity metrics were used.     

Estimation of storage time of yogurt with artificial neural network modeling

Changes in the physical, chemical, and microbiological structure of yogurt determine the storage and shelf life of the product. In this study, microbial counts and pH values of yogurt during storage were determined at d 1, 7, and 14. Simultaneously, image processing of yogurt was digitized by using a machine vision system (MVS) to determine color changes during storage, and the obtained data were modeled with an artificial neural network (ANN) for prediction of shelf life of set-type whole-fat and low-fat yogurts. The ANN models were developed using back-propagation networks with a single hidden layer and sigmoid activation functions. The input variables of the network were pH; total aerobic, yeast, mold, and coliform counts; and color analysis values measured by the machine vision system. The output variable was the storage time of the yogurt. The modeling results showed that there was excellent agreement between the experimental data and predicted values, with a high determination coefficient (R² = 0.9996) showing that the developed model was able to analyze nonlinear multivariant data with very good performance, fewer parameters, and shorter calculation time. The model might be an alternative method to control the expiration date of yogurt shown in labeling and provide consumers with a safer food supply.     

Bioavailability models for predicting copper toxicity to freshwater green microalgae as a function of water chemistry

We investigated whether an earlier-developed bioavailability model for predicting copper toxicity to growth rate of the freshwater alga Pseudokirchneriella subcapitata could be extrapolated to other species and toxicological effects (endpoints). Hardness and dissolved organic carbon did not significantly affect the toxicity of the free Cu2+ ion to P. subcapitata (earlier study) and Chlorella vulgaris(this study), but a higher pH resulted in an increased toxicity for both species. Regression analysis showed significant linear relationships between ECxpCu (= "effect concentration" that produces x% adverse effect, expressed as pCu = -log of the Cu2+ activity) and pH. By linking these regression models with a geochemical metal speciation model, dissolved copper concentrations that elicit a given adverse effect (EC(X)dissolved) can be predicted. Within the pH range investigated (5.5-8.7), slopes of the linear EC(X)pCu vs pH regression models varied between 1.301 and 1.472 depending on the species and the effect level (10% or 50%) considered. In a statistical sense these slopes were all significantly different from one another (p < 0.05), suggesting that this empirical regression model does not yet capture the full complexity of toxicological copper bioavailability to algae. However, we demonstrated that regression models with an "average" slope of 1.354 had predictive power very similar to those of regression models with species and effect-specific slopes. Additionally, the "average" regression model was further successfully validated for other species (Chlamydomonas reinhardtii and Scenedesmus quadricauda) and for different toxicological effects/endpoints (growth rate, biomass yield, and phosphorus uptake rate). For all these toxicity datasets effect concentrations of copper could be predicted with this "average" model by errors of less than a factor of 2 in 94-100% of the cases. The success of this "average" model suggests the possibility that the pH-based linear regression model may form a sound conceptual basis for modeling the toxicological bioavailability of copper to green algae in regulatory assessments, although a full mechanistic understanding is lacking and should be the focus of future studies.     

傅立叶近红外光谱法测定桃果中果胶含量

目的 基于傅里叶近红外光谱(Fourier transform near infrared)检测桃果中果胶含量的研究。方法 近红外光谱采集样品利用两个品种的桃,探究光谱预处理对建模的影响,建模采用偏最小二乘法(PLS)以及主成分回归(PCR)方法,模型的评价标准采用建模相关系数(R_C)、建模均方偏差(RMSE_C)、预测相关系数(R_P)、预测均方偏差(RMSE_P)。结果 两个品种的近红外光谱图和果胶含量无明显差异(P＞0.05),采用标准正态变量变换(SNV)和多元散射校正(MSC)对原始光谱的光程进行选择,所得建模结果影响基本一致,合适光谱数据格式以及平滑处理,能提高PLS和PCR模型的预测精度和稳定性。综合得出模型最佳是利用PLS方法建模并采用MSC/SNV结合一阶导数和 Savitzky-Golay (S-G)平滑对近红外光谱图进行预处理,评价参数分别为R_C=0.7795、R_P=0.7545、RMSE_C=0.0933、RMSE_P=0.0534和R_C=0.7800、R_P=0.7530、RMSE_C=0.0932、RMSE_P=0.0534。结论 该方法为利用近红外建模快速检测桃果中果胶含量提供重要依据。     

Development of an integrated model for heat transfer and dynamic growth of Clostridium perfringens during the cooling of cooked boneless ham

Numerous small meat processors in the United States have difficulties complying with the stabilization performance standards for preventing growth of Clostridium perfringens by 1 log10 cycle during cooling of ready-to-eat (RTE) products. These standards were established by the Food Safety and Inspection Service (FSIS) of the US Department of Agriculture in 1999. In recent years, several attempts have been made to develop predictive models for growth of C. perfringens within the range of cooling temperatures included in the FSIS standards. Those studies mainly focused on microbiological aspects, using hypothesized cooling rates. Conversely, studies dealing with heat transfer models to predict cooling rates in meat products do not address microbial growth. Integration of heat transfer relationships with C. perfringens growth relationships during cooling of meat products has been very limited. Therefore, a computer simulation scheme was developed to analyze heat transfer phenomena and temperature-dependent C. perfringens growth during cooling of cooked boneless cured ham. The temperature history of ham was predicted using a finite element heat diffusion model. Validation of heat transfer predictions used experimental data collected in commercial meat-processing facilities. For C. perfringens growth, a dynamic model was developed using Baranyi's nonautonomous differential equation. The bacterium's growth model was integrated into the computer program using predicted temperature histories as input values. For cooling cooked hams from 66.6 degrees C to 4.4 degrees C using forced air, the maximum deviation between predicted and experimental core temperature data was 2.54 degrees C. Predicted C. perfringens growth curves obtained from dynamic modeling showed good agreement with validated results for three different cooling scenarios. Mean absolute values of relative errors were below 6%, and deviations between predicted and experimental cell counts were within 0.37 log10 CFU/g. For a cooling process which was in exact compliance with the FSIS stabilization performance standards, a mean net growth of 1.37 log10 CFU/g was predicted. This study introduced the combination of engineering modeling and microbiological modeling as a useful quantitative tool for general food safety applications, such as risk assessment and hazard analysis and critical control points (HACCP) plans.     

Vulnerability of shallow groundwater and drinking-water wells to nitrate in the United States

Two nonlinear models were developed at the national scale to (1) predict contamination of shallow ground water (typically < 5 m deep) by nitrate from nonpoint sources and (2) to predict ambient nitrate concentration in deeper supplies used for drinking. The new models have several advantages over previous national-scale approaches. First, they predict nitrate concentration (rather than probability of occurrence), which can be directly compared with water-quality criteria. Second, the models share a mechanistic structure that segregates nitrogen (N) sources and physical factors that enhance or restrict nitrate transport and accumulation in ground water. Finally, data were spatially averaged to minimize small-scale variability so that the large-scale influences of N loading, climate, and aquifer characteristics could more readily be identified. Results indicate that areas with high N application, high water input, well-drained soils, fractured rocks or those with high effective porosity, and lack of attenuation processes have the highest predicted nitrate concentration. The shallow groundwater model (mean square error or MSE = 2.96) yielded a coefficient of determination (R(2)) of 0.801, indicating that much of the variation in nitrate concentration is explained by the model. Moderate to severe nitrate contamination is predicted to occur in the High Plains, northern Midwest, and selected other areas. The drinking-water model performed comparably (MSE = 2.00, R(2) = 0.767) and predicts that the number of users on private wells and residing in moderately contaminated areas (>5 to < or =10 mg/L nitrate) decreases by 12% when simulation depth increases from 10 to 50 m.     

Simultaneous Determination of Danofloxacin and Flumequine in Milk Based on Fluorescence Spectroscopy and Chemometrics Tools

The aim of this paper was to develop a rapid screening method to determine danofloxacin (DANO) and flumequine (FLU) in milk by fluorescence spectroscopy combined with three different chemometric tools. In this study, 2-D fluorescence data and multivariate calibration based on a partial least squares discriminant analysis (PLS-DA) regression were combined to simultaneously qualify and quantify DANO and FLU concentrations in commercial ultra-high-temperature (UHT) sterilized and pasteurized milk. Calibration sets based on the UHT whole milk from brand A were built and performed using a partial least squares (PLS) regression after deproteinization. Prediction sets based on 13 types of milk were analyzed using principal component analysis (PCA), principal PLS-DA, and PLS regression models. The multivariate calibration models were better able to determine the DANO and FLU concentrations than the univariate models, and these models could be applied to other types of milk. In contrast to the PLS-DA, which had good sensitivity and specificity, the PCA yielded less satisfactory results. In the quantitative analysis, the recoveries of the two analytes were reasonable and the root mean square error of prediction was within the acceptable range. The relative standard deviations of the predicted DANO and FLU concentrations on the various testing days were 9.2 and 6.2 %, respectively, demonstrating that the analytical method had a good reproducibility.     

激光近红外结合SVM的花生油掺伪定性定量分析

利用激光近红外技术结合支持向量机(support vectormachines,SVM)对花生油掺伪进行定性和定量分析。使用激光近红外光谱仪采集188个掺入餐饮废弃油、大豆油、玉米油以及菜籽油的花生油样品光谱图。结果表明,建立的SVC分类模型均能实现100%的预测准确率,但经提取波长后的模型的变量变少,由全波段的451个波长数减少为136个。建立的SVR回归模型也能准确预测花生油中掺伪油的含量,其中非全波段模型参与建模变量变少,由451个降低到66个,预测精度也更高,校正集和测试集相关系数分别达到99.88%、99.90%,均方根误差都低于6.99E-4。由此可知,特征波长提取方法不仅可以减少建模变量,提高建模效率,也能够提高模型的预测能力。结果表明,运用激光近红外结合SVM可以实现花生油掺伪油脂的定性和定量分析。     

Classification of intramural metastases and lymph node metastases of esophageal cancer from gene expression based on boosting and projective adaptive resonance theory

Esophageal cancer is a well-known cancer with poorer prognosis than other cancers. An optimal and individualized treatment protocol based on accurate diagnosis is urgently needed to improve the treatment of cancer patients. For this purpose, it is important to develop a sophisticated algorithm that can manage a large amount of data, such as gene expression data from DNA microarrays, for optimal and individualized diagnosis. Marker gene selection is essential in the analysis of gene expression data. We have already developed a combination method of the use of the projective adaptive resonance theory and that of a boosted fuzzy classifier with the SWEEP operator denoted PART-BFCS. This method is superior to other methods, and has four features, namely fast calculation, accurate prediction, reliable prediction, and rule extraction. In this study, we applied this method to analyze microarray data obtained from esophageal cancer patients. A combination method of PART-BFCS and the U-test was also investigated. It was necessary to use a specific type of BFCS, namely, BFCS-1,2, because the esophageal cancer data were very complexity. PART-BFCS and PART-BFCS with the U-test models showed higher performances than two conventional methods, namely, k-nearest neighbor (kNN) and weighted voting (WV). The genes including CDK6 could be found by our methods and excellent IF-THEN rules could be extracted. The genes selected in this study have a high potential as new diagnosis markers for esophageal cancer. These results indicate that the new methods can be used in marker gene selection for the diagnosis of cancer patients.