ABSTRACT
Although many efforts have contributed to improve our knowledge of molecular pathogenesis about multiple myeloma (MM), the role and significance of microRNAs and long noncoding RNAs in MM cells, along with the core mechanism remains virtually absent. The mRNA levels of miR-125b and MALAT1 in MM cell lines were detected by qRT-PCR. The influence of Lenti-Sh-miR-125b on cell viability and the Notch-1 pathway-related proteins were assessed by MTT method and western blot, respectively. We also investigated the regulation effect between MALAT1 and Notch1 pathway. Moreover, the connection between Notch1 signaling and MM cell growth was discussed in-depth. The reverse effect of pcDNA-Notch1 on the cell viability and Notch-1 pathway proteins induced by Si-MALAT1 was also studied. Furthermore, miR-125b overexpressing MM cell lines were injected subcutaneously into nude mice. MiR-125b and MALAT1 were inversely expressed in MM cell lines. Lenti-Sh-miR-125b inhibited the expression of MALAT1 and Notch-1 protein. Binding sites were confirmed between miR-125b and MALAT1, and silencing MALAT1 did not alter the expression of Notch-1. The apoptosis rate was increased and the survival rate was decreased obviously in GSI XII (targeted cleavage of Notch-1 receptor) group, along with the inhibited Notch1 and HES1 proteins. Moreover, the decreased cell viability and Notch-1 pathway proteins induced by Si-MALAT1 could be reversed by pcDNA-Notch1. Lenti-Sh-miR-125b promoted survival and decreased Notch1 and HES1 proteins levels, while this effect was reversed by si -MALAT1. MiR-125b regulated MALAT1 expression via Notch1 signaling pathway to regulate cell growth, thus participating in the occurrence and progression of MM, which functioned as a therapeutic target for tracking MM.
Subject(s)
Gene Expression Regulation, Neoplastic , MicroRNAs/metabolism , Multiple Myeloma/genetics , RNA, Long Noncoding/metabolism , Animals , Apoptosis/genetics , Cell Line, Tumor , Cell Survival/genetics , Disease Progression , Humans , Mice , Mice, Inbred BALB C , Mice, Nude , Multiple Myeloma/pathology , RNA, Long Noncoding/genetics , RNA, Messenger/metabolism , RNA, Small Interfering/metabolism , Receptor, Notch1/metabolism , Signal Transduction/genetics , Xenograft Model Antitumor AssaysABSTRACT
Long non-coding RNAs (lncRNAs) can participate in the pathological process of multiple myeloma (MM) via regulation of specific gene expression and function. This research aimed to study the role of MALAT-1 and the underlying mechanism in MM. In this study, the expression of MALAT-1 and HMGB1 protein in the bone marrow mononuclear cells from MM patients at different stages and in MM cell lines was determined by qRT-PCR and western blot, respectively. The endogenous expression of MALAT-1 and HMGB1 was modulated using lentivirus vectors transfection. CHX chase assay and RIP analyses were performed to explore the interaction between MALAT-1 and HMGB1 in MM. Nude mouse xenograft was made and used for in vivo experiment study. The expression of MALAT-1 and HMGB1 in the bone marrow mononuclear cells from patients with untreated multiple myeloma was dramatically increased, as well as in MM cell lines, KM3 and U266; while MALAT-1 expression and HMGB1 protein level both decreased significantly in complete remission patients. Furthermore, MALAT-1 knockdown facilitated the degradation of HMGB1 at the post-translational level via increase of the ubiquitination of HMGB1 in MM cells. MALAT-1 was shown to promote autophagy in MM through upregulation of HMGB1. In vivo, MALAT-1 knockdown could inhibit tumor growth significantly in tumor-bearing mice and reduced the protein expressions of HMGB1, Beclin-1, and LC3B in tumor tissues. LncRNA MALAT-1 increases the expression level of HMGB1 in MM thereby promotes autophagy resulting in the inhibition of apoptosis. J. Cell. Biochem. 118: 3341-3348, 2017. © 2017 Wiley Periodicals, Inc.
Subject(s)
Apoptosis , Autophagy , HMGB1 Protein/biosynthesis , Multiple Myeloma/metabolism , Neoplasm Proteins/biosynthesis , RNA, Long Noncoding/metabolism , RNA, Neoplasm/metabolism , Animals , Cell Line, Tumor , Female , HMGB1 Protein/genetics , Humans , Male , Mice , Mice, Inbred BALB C , Mice, Nude , Middle Aged , Multiple Myeloma/genetics , Multiple Myeloma/pathology , Neoplasm Proteins/genetics , RNA, Long Noncoding/genetics , RNA, Neoplasm/geneticsABSTRACT
In order to expand the application range of the model for a single kind of fruits in the portable near infrared instrument, this paper comes up with a new method for the soluble solid content (SSC) model transfer between different kinds of fruits. This method is focusing on the idea of model transfer between different instruments. Based on the similar physical and chemical properties of apples, peaches and pears, such as the range of SSC content, fruit size and the thickness of peel, a simple Slope/Bias algorithm is applied to the transfer of apple SSC partial least square (PLS) model. After that, it can be used to predict pear & peach SSC value with very little extra samples. It's more convenient and costs less by using this method. For pear samples, by using extra 35 standard samples to transfer apple SSC model, RMSEP reduced from 1.009 °Brix to 0.565 °Brix. For peaches, extra 40 standard samples led to a significant reduce of RMSEP from 1.726 °Brix to 0.677 °Brix after model transfer. To validate the feasibility of this model transfer method, both pear and peach SSC models were tested using the same Slope/Bias algorithm model transfer respectively. A pear SSC model was firstly set up and then transferred with Slope/Bias method. Taking 30 standard apples as samples, RMSEP value reached 0.597°Brix, while taking 40 standard peaches as samples, RMSEP value reached 0.689°Brix. The peach SSC model was transferred in the same way. For apples, using 35 standard samples, RMSEP value reached 0.654°Brix, and for pears, using 30 standard samples, RMSEP value reached 0.439°Brix. These results show that slope/bias algorithm can be used to transfer model between similar kinds of fruits such as apples, pears and peaches. The paper provides innovative ideas for the model transfer among similar kinds of materials, so that the portable near infrared instruments can be used more conveniently and widely.
ABSTRACT
To ensure the material safety of dairy products, visible (Vis)/near infrared (NIR) spectroscopy combined with che- mometrics methods was used to develop models for fat, protein, dry matter (DM) and lactose on-site evaluation. A total of 88 raw milk samples were collected from individual livestocks in different years. The spectral of raw milk were measured by a porta- ble Vis/NIR spectrometer with diffused transmittance accessory. To remove the scatter effect and baseline drift, the diffused transmittance spectra were preprocessed by 2nd order derivative with Savitsky-Golay (polynomial order 2, data point 25). Changeable size moving window partial least squares (CSMWPLS) and genetic algorithms partial least squares (GAPLS) meth- ods were suggested to select informative regions for PLS calibration. The PLS and multiple linear regression (MLR) methods were used to develop models for predicting quality index of raw milk. The prediction performance of CSMWPLS models were similar to GAPLS models for fat, protein, DM and lactose evaluation, the root mean standard errors of prediction (RMSEP) were 0.115 6/0.103 3, 0.096 2/0.113 7, 0.201 3/0.123 7 and 0.077 4/0.066 8, and the relative standard deviations of prediction (RPD) were 8.99/10.06, 3.53/2.99, 5.76/9.38 and 1.81/2.10, respectively. Meanwhile, the MLR models were also cal- ibrated with 8, 10, 9 and 7 variables for fat, protein, DM and lactose, respectively. The prediction performance of MLR models was better than or close to PLS models. The MLR models to predict fat, protein, DM and lactose yielded the RMSEP of 0.107 0, 0.093 0, 0.136 0 and 0.065 8, and the RPD of 9.72, 3.66, 8.53 and 2.13, respectively. The results demonstrated the usefulness of Vis/NIR spectra combined with multivariate calibration methods as an objective and rapid method for the quality evaluation of complicated raw milks. And the results obtained also highlight the potential of portable Vis/NIR instruments for on-site assessing quality indexes of raw milk.
Subject(s)
Milk/chemistry , Spectroscopy, Near-Infrared , Algorithms , Animals , Calibration , Dietary Fats/analysis , Lactose/analysis , Least-Squares Analysis , Linear Models , Milk Proteins/analysisABSTRACT
In the present research, synchronous fluorescence technique was used for qualitative and quantitative detection of re- constituted milk mixed into two kinds of milk samples, raw milk and pasteurized milk, respectively. The total accuracy of sample was used to evaluate the performance of the qualitative discrimination models. The correlation coefficient (r), the root mean square error of correction (RMSEC) and the root mean square error of prediction (RMSEC) were used to evaluate the perform- ance of the quantitative analysis models. The constant wavelength difference (Δλ) between the excitation and emission scanning was determined to be 80 nm from three-dimensional fluorescence spectrum of milk. The total discrimination accuracy was 100% by partial least squares discrimination analysis (PLS-DA) for raw milk, pasteurized milk and reconstituted milk samples. When checking whether the raw milk and pasteurized milk were mixed with reconstituted milk, the total accuracy of calibration samples was 100% and the accuracy of prediction samples was 75% and 81.25%, respectively. The effects of qualitative discrimination models were satisfactory. The PLS regression was used for quantitative analysis of the reconstituted milk content mixed in raw milk and pasteurized milk. The correlation coefficients of actual values versus predicted values were 0.911 2 and 0.911 2, re-spectively. The RMSEC was 0.042 2 and 0.0384, respectively. The RMSEP was 0.054 8 and 0.057 5, respectively. The cor- relation coefficients of quantitative analysis models could reach up to 0.9. The results showed that synchronous fluorescence technology could be applied for rapid detection of reconstituted milk mixed in fresh milk
Subject(s)
Fluorescence , Food Contamination , Milk/chemistry , Animals , Calibration , Least-Squares AnalysisABSTRACT
Dividing watermelons into two categories as not complete mature and fully mature by cluster analyzing the 10 indicators associated with maturity, the two modeling methods PCADA and PLSDA were used, and through the near-infrared spectroscopy, the maturity of small watermelon fruit JINGXIU was qualitatively determined. The PCADA model is the best. Modeling at the top position is better than that of the equatorial parts of the melon. The two models both have a miscarriage of justice, and exists the same sample with a miscarriage of justice. Fruit samples of different physical and chemical composition and structure will have an impact on the spectral information, resulting in miscarriage of justice. Near-infrared diffuse transmittance technique can get better results in detection of small watermelon maturity. But the prediction model should be established to select the appropriate parts of the spectrum acquisition and modeling methods.
Subject(s)
Citrullus , Fruit , Spectroscopy, Near-InfraredABSTRACT
In order to identify American ginseng and panax ginseng samples accurately and rapidly, the authors acquired the NIR spectra of the samples' cross-sections. Then the spectra were respectively analyzed according to the samples' physical structure factors and chemical factors. The authors selected appropriate bands and built a physical factor leading model, a chemical factors leading model as well as a comprehensive factor model. The authors found that all the three models' discriminant rates were above 96 percents, which can meet the needs of the rapid detection of raw Chinese medicinal crop materials. While the physical factors model had a simple operation, the discriminant rate was relatively low. The chemical factors model' discriminant rate was higher, but the computation is much more complex. Among the three models, the mixed factor model had the best result with the highest discrimination rate (100 percents) and a smaller number of principal components (4). The effect was the most ideal. It proved that physical factors play an important part in NIR modeling. The cross section method is accurate and convenient which can be used in the quality control in enterprise, realizing the rapid screening of the medicine raw materials.
Subject(s)
Panax/chemistry , Panax/classification , Spectroscopy, Near-Infrared , Drugs, Chinese Herbal/analysis , Models, Theoretical , Quality ControlABSTRACT
The deterioration and shell thickness of walnut were studied using the terahertz time domain spectroscopy. Firstly, the THz spectra of moth-eaten, moldy and normal walnuts were compared, and the bad walnuts were properly rejected due to the differences of absorption peaks. Secondly, the transmission-type and reflection-type terahertz time domain spectroscopy system was used simultaneously, and a new formula to calculate shell thickness of walnut was built in the THz system. Then the authors measured the shell thickness based on the detectable refractive index of walnut, and the relative error was 3.7%. Consequently, the quality of walnut was evaluated nondestructively according to physical and chemical indicators from walnut THz spectra respectively.
Subject(s)
Food Analysis/methods , Juglans , Nuts , Terahertz Spectroscopy/methods , Quality ControlABSTRACT
Near infrared diffuse reflectance spectroscopy calibrations of fat, protein and DM in raw milk were studied with partial least-squares (PLS) regression using portable short-wave near infrared spectrometer. The results indicated that good calibrations of fat and DM were found, the correlation coefficients were all 0.98, the RMSEC were 0.187 and 0.217, RMSEP were 0.187 and 0.296, the RPDs were 5.02 and 3.20 respectively; the calibration of protein needed to be improved but can be used for practice, the correlation coefficient was 0.95, RMSEC was 0.105, RMSEP was 0.120, and RPD was 2.60. Furthermore, the measuring accuracy was improved by analyzing the correction relation of fat and DM in raw milk This study will probably provide a new on-site method for nondestructive and rapid measurement of milk.
Subject(s)
Fats/analysis , Milk/chemistry , Proteins/analysis , Spectroscopy, Near-Infrared/methods , AnimalsABSTRACT
Spectral data compression and informative variable selection are the research focus on the application of NIR, which enable to simplify the model and improve the accuracy of prediction. The research used the pretreatment methods such as the second derivative, normalization and orthogonal signal correction (OSC) to filter irrelevant array according to the concentration of soluble solid content (SSC) based on the Vis/NIR spectroscopy of apricot. SCMWPLS was used to select 880, 894-910 and 932 nm as the regions for constructing prediction PLS model with correlation coefficient (R) of 0.920, standard error of calibration (SEC) of 0.454 and standard error of prediction (SEP) of 0.470 for SSC. Besides, after conducting an independent run for 100 times, GA obtained the regression variables as 888 and 900 nm according to the higher frequency of selection to set up GA. MLR prediction model, and the R, SEC and SEP were 0.905, 0.488 and 0.459 respectively. The results of the two modeling methods are both better than those of full-region PLS model. This demonstrates that OSC enables to filter irrelevant signal array according to the concentration of SSC and reduce the latent variables used for modeling. Also, SCMWPLS and GA can identify the optimal combination of information variables. These methods have a universal significance on building NIR express analysis model with low dimension and high precision.
ABSTRACT
Near infrared spectroscopy combined with pattern recognition techniques were applied to develop a method of fast and nondestructive discrimination between Chinese ginseng and American ginseng. A total of 90 representative ginseng samples including root, fiber and powder were collected. NIR spectra of the samples were obtained directly with wrapped polyethylene packing film. MSC and first derivative were performed after the elimination of notable packing film absorbance in raw spectra. Then the informative wave bands were chosen by moving window partial least-squares regression method. PLS-DA, PCA-DA and SVM discrimination models were founded and their results were compared. SVM was proven to be the most effective method with 100% accurate identification rate for validation set. It indicates that the method founded is precise and convenient and can be practically used in practice for quality control and fast screening of raw herb materials.
Subject(s)
Panax/classification , Spectroscopy, Near-Infrared , Least-Squares Analysis , Quality ControlABSTRACT
The feasibility of using efficient selection of variables in Vis/NIR for a rapid and conclusive determination of fruit inner qualities such as soluble solids content (SSC) of plums was investigated. A new strategy was proposed in the present paper, i. e. two-stage variable selection using the backward interval partial least squares (BiPLS) combined with genetic algorithm (GA). Firstly, it splits the whole spectral region into equidistant sub-regions and then develops all BiPLS regression models, and the informative regions which are used to constructed PLS models with the lowest error can be located. Secondly, GA method is used to select variable in these informative regions, which are used for regression variables of MLR model. The Vis/NIR spectra containing 225 individual data points were processed by Savizky-Golay filter smoothing and second-order derivative, and 9 sub-regions were selected by BiPLS procedure when the spectra were divided into 25 sub-regions. The optimal 12 variables, which were the output of the GA procedure, were selected by the higher occurrence frequency while the GA procedure ran 100 times. In order to simplify the multiple linear regression (MLR) modeling, the wavelength variables with the maximum occurrence frequency were chosen when the adjacent wavelengths were selected by GA. Finally, 638, 734, 752, 868, 910, 916 and 938 nm were used to build a MLR model. The results show that MLR model produced by BiPLS-GA performs well with correlation coefficients (R) of 0.984, root mean standard error of calibration (RMSEC) of 0.364 and root mean standard error of prediction (RMSEP) of 0.471 for SSC, which outperforms models using stepwise regression analysis (SRA). This work proved that the BiPLS-GA could determine optimal variables in Vis/NIR spectra and improve the accuracy of model.
ABSTRACT
The detection precision of soluble solids in apple fruit by near infrared reflectance (NIR) spectroscopy was affected by sample temperature. The NIR technique needs to be able to compensate for fruit temperature fluctuations. In the present study, it was observed that the sample temperature (2-42 degrees C) affects the NIR spectrum in a nonlinear way. The temperature model was built with R2 = 0.985, RMSEC = 1.88, and RMSEP = 2.32. When no precautions are taken, the error in the SSC reading may be as large as 2.55% degrees Brix. Two techniques were found well suited to control the accuracy of the calibration models for soluble solids with respect to temperature fluctuations, such as temperature variable--eliminating calibration model and global robust calibration model to cover the temperature range. And an improved genetic algorithms (GAs) was used to implement an automated variables selection procedure for use in building multivariate calibration models based on partial least squares regression (PLS). The two compensation methods were found to perform well with RMSEP1 = 0.72/0.69 and RMSEP2 = 0.74/0.68, respectively. This work proved that the compensation techniques could emend the temperature effect for NIR spectra and improve the precision of models for apple SSC by NIR.
ABSTRACT
The robust NIRS model must be developed by the representative samples and precise chemical values, taking much of work. To reduce the calibration work, the present paper explored the NIRS model developed using ethanol liquor to predict ethanol of the wine samples. The authors used the gene arithmetic (GA) method to select the calibration region (2 245-2 320 nm) which has relatively high correlation with the consistency of ethanol in ethanol liquor and has little interfere by other components in wine. To remove the systematic error between the calibration set of ethanol liquor and the prediction set of turbid vinous ferment liquid, according to the method of slope/bias, the authors selected 21 samples in prediction set which can represent the range of consistency of vinous ferment liquid to revise the ethanol model in order to predict the remaining wine samples well. After the calculation, the authors obtained the bias and the slope to be 0.523 3 and 0.980 8, respectively. Then we predicted the other turbid samples of wine using the ethanol liquor model after being revised by the slope/bias method. And the prediction model for the ethanol of turbid samples was developed, with r, RPD and RSD for the prediction model for ethanol of samples being 0.99%, 11.71% and 3.11%, respectively, indicating that the ethanol liquor model is robust and can serve as the model of vinous ferment liquid to detect the ethanol of the wine. So this method can largely reduce the calibration work during the NIR calibration process, and has the practical feasibility and application value.
ABSTRACT
The present paper presents a new NIR analysis method with partial least square regression (PLS) and artificial neural network (ANN) to improve the prediction precision of the protein model for milk powder. First, an efficient method named region selecting by genetic algorithms (RS-GA) was used to select the calibration region, and then the GA-PLS model was made to predict the linear part of the protein content in milk powder. And then in the region selected by RS-GA method, principal component analysis (PCA) was calculated. The principal components were taken as the input of ANN model. The remnant values by subtracting the standard values and the GA-PLS validation values were regarded as the output of ANN. The ANN model was made to predict the nonlinear part of the protein content. The final result of the model was the addition of the two model's validation values, and the root mean squared error of prediction (RMSEP) was used to estimate the mixed model. A full region PLS model (Fr-PLS) was also made, and the RMSEP of the Fr-PLS, GA-PLS and GA-PLS+PC-ANN model was 0.511, 0.440 and 0.235, respectively. The results show that the prediction precision of the protein model for milk powder was largely improved when adding the nonlinear port in the NIR model, and this method can also be used for other complex material to improve the prediction precision.
Subject(s)
Algorithms , Milk Proteins/analysis , Milk Substitutes/chemistry , Models, Statistical , Neural Networks, Computer , Principal Component Analysis , Least-Squares Analysis , PowdersABSTRACT
Genetic algorithm is widely used in NIRS data optimization, which is not limited by searching space. The region selecting by genetic algorithms (R-SGA)was applied in building calibration model of soluble solid content (SSC)of Pyrus pyrifolia, and the number of variables used to build calibration was further reduced from 2 075 to 690 in all of 3 models. Studies were performed to build GA-PLS models by different R-SGA latent variables, and the optimal R-SGA latent variables of Hosui, Wonhwang and Whangkeumbae pear were 10, 12 and 16, respectively. The R-SGA procedure was found to perform well (RMSEP = 0.608 and 0.524 for Hosui and Whangkeumbae pear respectively), leading to calibration models that significantly outperform those based on full-spectrum analyses (RMSEP = 0.632, 0.540). The prediction precision of GA-PLS models was similar to FULL-PLS for Wonhwang pear, with RMSEP of 0.610/0.595. In addition, the selected regions from R-SGA methods were used to build mixed model of 3 Pyrus pyrifolia varieties. The results indicated that the prediction precision of GA-PLS model was close to that of the full spectrum model, with RMSEP of 0.641 and 0.645, respectively. This work proved that the R-SGA could find optimal values for several disparate variables associated with the calibration model and that the PLS procedure could be integrated into the objective function driving the optimization, and it was feasible to build a universal model of different Pyrus pyrifolia varieties.
Subject(s)
Algorithms , Models, Theoretical , Pyrus/chemistry , Calibration , Genetics , Least-Squares Analysis , SolubilityABSTRACT
An improved genetic algorithm was used to implement an automated wavelength selection procedure for use in building multivariate calibration models based on partial least squares regression (PLS). The region selecting by genetic algorithms (R-SGA) was applied in building calibration model of soluble solid content (SSC) of Western pear, and the numbers of latent variables used to build calibration model were further reduced. The Fourier transform near infrared reflectance (FT-NIR) spectra were processed by GA after MSC or SNV, and four PLS calibration models were built by using the optimal combinations of these sub-regions. Meanwhile, the full region selecting PLS (Fr-PLS) models were developed. The R-SGA models variables were 434, 496, 310 and 496, for Early Red Comice, Wujiuxiang, Cascade and Kang Buddha, respectively. Despite the complexity of the spectral data, the R-SGA procedure was found to perform well (RMSEP = 0.428, 0.567 for Early Red Comice and Kang Buddha, respectively), leading to calibration models that significantly outperform those based on full-spectrum analyses (RM-SEP = 0.518, 0.633). The prediction precision of GA-PLS models was similar to that of Fr-PLS for Wujiuxiang and Cascade, with RMSEP of 0.696/0.694 and 0.425/0.421 respectively. This work proved that the R-SGA could find optimal values for several disparate variables associated with the calibration model and that the PLS procedure could be integrated into the objective function driving the optimization.
Subject(s)
Algorithms , Pyrus/chemistry , Spectroscopy, Fourier Transform Infrared/statistics & numerical data , Calibration , Genetics , Least-Squares Analysis , SolubilityABSTRACT
It is urgent to develop a quick and precise method for the discrimination of the internal quality of apple. Vis/NIR spectroscopy combined with multivariate analysis after the appropriate spectral data pre-treatment has been proved to be a very powerful tool for judgment of objects that have very similar exterior properties. In the present study, peak area discriminant analysis (PADA), principal component analysis discriminant analysis (PCADA) and partial least squares discriminant analysis (PLSDA) were applied to classify apples with different internal properties such as brownherat and watercore. Energy spectra were processed using MSC or one-order derivative, and three models using PADA, PCADA and PLSDA were built, respectively. The accuracy rates of prediction for brownheart apple were 100%, for watercore apple were 79.6%, 95.0% and 96.7%, and for natural apple were 88.4%, 98.2% and 98.8%, respectively. The PLSDA model was better than the others remarkably. And the overall correct ratio of PLSDA was 98.1%, with RMSEC = 0.449 and RMSEP = 0.392. The results in the present study show that Vis/NIR spectroscopy together with chemometrics techniques could be used to differentiate brownheart and watercore apple, which offers the benefit of avoiding time-consuming, costly and sensory analysis.
Subject(s)
Food Contamination/analysis , Malus/chemistry , Spectroscopy, Near-Infrared , Algorithms , Discriminant Analysis , Least-Squares Analysis , Principal Component Analysis/methodsABSTRACT
A new non-destructive and rapid method was developed to discriminate varieties of corn seeds. The method is based on near-infrared reflectance spectroscopy (NIRS) and artificial neural network (ANN). The corn seeds used for this study involved four varieties: Gaoyou115, Nongda368, Nongda108 and Nongda4967. After collecting the near-infrared reflectance spectrum of each single seed in the range between 1000 and 2632 nm, the principal component analysis (PCA) was used to compress the NIR spectra, which had been preprocessed with Savitky-Golay and multiplicative scatter correction (MSC). The analysis results showed that the cumulate reliabilities of PC1 to PC8 (the first eight principal components) were 99.602%. A three-layer back-propagation neural network (BPNN) was developed for classification, which was trained by the Levenberg-Marquard algorithm to improve the network training speed and efficiency. The LMBP was activated by the sigmoid function, and normalization of targets was used to get the best discrimination result of network. The first eight principal components of the samples were applied as LMBPNN inputs, and the values of the type of corn seeds were applied as the outputs. In this model, 120 kernels were used as the training data set and 40 kernels were used as the test data set. Calculation results showed that the distinguishing rate of the four corn seed varieties was 95%. This model is reliable and practicable. The results demonstrated that this identification method was rapid and non-destructive, and could be used for classification.
Subject(s)
Neural Networks, Computer , Seeds/chemistry , Spectroscopy, Near-Infrared , Zea mays/chemistry , Algorithms , Models, Chemical , Principal Component Analysis/methods , Spectroscopy, Near-Infrared/methodsABSTRACT
In the present paper, the effects of different milk processing (homogenizing process, pasteurization) on the milk NIR spectra were discussed. It was found that the raw milk and processed milk show significant difference in the 1 890 nm region, which can be used not only to identify the processed milk, but also to offer the basic theory for NIRS in the quality control researches of milk. The absorbance sharply reduced when the liquid milk was treated by a homogenizer, but the absorbance increased after pasteurization. Raw milks absorbance shows a downtrend in the whole region of spectra with increasing pressure. The changes in fat globules structure finally result in absorbance decline. The commercial milk including remade milk was taken for example to discuss the mechanism of detection. The discriminate analysis calibration was developed by SIMCA method and the accuracy of detection is 98.1% for identifying the reconstituted milk in pasteurized milk between 1 800 and 2 200 nm with the pretreatment method of second derivatived and Norris 5.5.