Research article

Validation of a rapid test to dose SO2 in vinegar

  • Sulfur dioxide is generally used in wine and vinegar production. It is employed to decrease the bacteria' growth, improve the wines' aroma (since it supports the extraction of polyphenols during maceration), protect the wines from chemical oxidation and the musts from chemical and enzymatic oxidation (blocking free radicals and oxidase enzymes such as tyrosinase and laccase). The composition and storage conditions (i.e., pH, temperature, and alcohol levels) affect oenological results. In various countries, competent authorities have imposed legal limits since it can have toxic effects on humans. It is crucial to dose SO2 levels to allow vinegar production and compliance with legal limits. The iodometric titration named "Ripper test" is the legal method used to dose it in vinegar. In this work, an automatized colorimetric test was validated using the international guidelines ISO/IEC (2017) to allow its use instead of the Ripper test. The test reliability was verified on white, red, and balsamic vinegar with low or high SO2 content. The automatized test showed linearity, precision, and reproducibility similar to the Ripper test, but the accuracy parameter was not respected for the vinegar with a low concentration of SO2. Therefore, the automatized colorimetric test can be helpful to dose SO2 in vinegar with high concentrations of SO2.

    Citation: Irene Dini, Antonello Senatore, Daniele Coppola, Andrea Mancusi. Validation of a rapid test to dose SO2 in vinegar[J]. AIMS Agriculture and Food, 2023, 8(1): 1-24. doi: 10.3934/agrfood.2023001

    Related Papers:

    [1] Ansheng Ye, Xiangbing Zhou, Kai Weng, Yu Gong, Fang Miao, Huimin Zhao . Image classification of hyperspectral remote sensing using semi-supervised learning algorithm. Mathematical Biosciences and Engineering, 2023, 20(6): 11502-11527. doi: 10.3934/mbe.2023510
    [2] Wenjun Xu, Zihao Zhao, Hongwei Zhang, Minglei Hu, Ning Yang, Hui Wang, Chao Wang, Jun Jiao, Lichuan Gu . Deep neural learning based protein function prediction. Mathematical Biosciences and Engineering, 2022, 19(3): 2471-2488. doi: 10.3934/mbe.2022114
    [3] Lili Jiang, Sirong Chen, Yuanhui Wu, Da Zhou, Lihua Duan . Prediction of coronary heart disease in gout patients using machine learning models. Mathematical Biosciences and Engineering, 2023, 20(3): 4574-4591. doi: 10.3934/mbe.2023212
    [4] Shun Li, Lu Yuan, Yuming Ma, Yihui Liu . WG-ICRN: Protein 8-state secondary structure prediction based on Wasserstein generative adversarial networks and residual networks with Inception modules. Mathematical Biosciences and Engineering, 2023, 20(5): 7721-7737. doi: 10.3934/mbe.2023333
    [5] Xiaoman Zhao, Xue Wang, Zhou Jin, Rujing Wang . A normalized differential sequence feature encoding method based on amino acid sequences. Mathematical Biosciences and Engineering, 2023, 20(8): 14734-14755. doi: 10.3934/mbe.2023659
    [6] Mingju Chen, Zhongxiao Lan, Zhengxu Duan, Sihang Yi, Qin Su . HDS-YOLOv5: An improved safety harness hook detection algorithm based on YOLOv5s. Mathematical Biosciences and Engineering, 2023, 20(8): 15476-15495. doi: 10.3934/mbe.2023691
    [7] Jingren Niu, Qing Tan, Xiufen Zou, Suoqin Jin . Accurate prediction of glioma grades from radiomics using a multi-filter and multi-objective-based method. Mathematical Biosciences and Engineering, 2023, 20(2): 2890-2907. doi: 10.3934/mbe.2023136
    [8] Natalya Shakhovska, Vitaliy Yakovyna, Valentyna Chopyak . A new hybrid ensemble machine-learning model for severity risk assessment and post-COVID prediction system. Mathematical Biosciences and Engineering, 2022, 19(6): 6102-6123. doi: 10.3934/mbe.2022285
    [9] Lu Yuan, Yuming Ma, Yihui Liu . Protein secondary structure prediction based on Wasserstein generative adversarial networks and temporal convolutional networks with convolutional block attention modules. Mathematical Biosciences and Engineering, 2023, 20(2): 2203-2218. doi: 10.3934/mbe.2023102
    [10] Hong Yuan, Jing Huang, Jin Li . Protein-ligand binding affinity prediction model based on graph attention network. Mathematical Biosciences and Engineering, 2021, 18(6): 9148-9162. doi: 10.3934/mbe.2021451
  • Sulfur dioxide is generally used in wine and vinegar production. It is employed to decrease the bacteria' growth, improve the wines' aroma (since it supports the extraction of polyphenols during maceration), protect the wines from chemical oxidation and the musts from chemical and enzymatic oxidation (blocking free radicals and oxidase enzymes such as tyrosinase and laccase). The composition and storage conditions (i.e., pH, temperature, and alcohol levels) affect oenological results. In various countries, competent authorities have imposed legal limits since it can have toxic effects on humans. It is crucial to dose SO2 levels to allow vinegar production and compliance with legal limits. The iodometric titration named "Ripper test" is the legal method used to dose it in vinegar. In this work, an automatized colorimetric test was validated using the international guidelines ISO/IEC (2017) to allow its use instead of the Ripper test. The test reliability was verified on white, red, and balsamic vinegar with low or high SO2 content. The automatized test showed linearity, precision, and reproducibility similar to the Ripper test, but the accuracy parameter was not respected for the vinegar with a low concentration of SO2. Therefore, the automatized colorimetric test can be helpful to dose SO2 in vinegar with high concentrations of SO2.



    Neurotoxins are toxic substances that are destructive to nerve tissue, such as AF64A, 6-hydroxydopamine, and kainic acid. In principle, the toxins act on the ion channels in the nerve-muscle junction, destroying cholinergic neurons, inhibiting the release of acetylcholine, blocking nerve-muscle conduction, causing muscle weakness, and thus making the muscle unable to contract. In severe cases, these toxins can cause suffocation and death. Neurotoxins can be classified into presynaptic and postsynaptic types according to their mechanism of action [1]. Presynaptic neurotoxins mainly act on the presynaptic membrane [2]; due to the specificity of enzyme activity, they typically block neuromuscular transmission and inhibit the release of neurotransmitters. The targets of postsynaptic neurotoxins are located in the postsynaptic membrane and can bind to acetylcholine receptors [3]. For example, β-methylamino-L-alanine, also known as BMAA, can damage motor neurons and has been implicated in Parkinson's syndrome. Cobra neurotoxin is a short-chain neurotoxin, the most important lethal component in cobra venom, a mainly postsynaptic neurotoxin. Because cobra venom neurotoxin is nonaddictive and non-drug resistant, it has broad prospects for persons with a drug addiction in detoxification. Therefore, the study of presynaptic and postsynaptic neurotoxins will contribute to the development of medicine, for example, to provide important clues for drug design [4,5,6].

    Neurotoxins are a type of protein, and while their structure and function can be correctly predicted through biochemical experiments, the work is time-consuming and expensive [7,8,9,10]. In the genome era, many biological sequences are available [11], giving us a variety of methods to predict protein structure and function [12,13,14,15]. The key to correctly predicting protein structure and function is how to analyze these features using computational methods. Therefore, we can use machine learning methods for protein type prediction [16]. Generally, the use of machine learning to predict biological sequences mainly includes the following steps: feature extraction, model construction, and performance evaluation [17,18,19,20,21,22]. In 2009, a diversity-based method of identifying presynaptic and postsynaptic neurotoxins was proposed. The algorithm is based on the composition of amino acids and pseudo-amino acids [23]. To further improve the accuracy of prediction, Hua Tang et al. proposed a new feature selection technique based on the principle of variance analysis (ANOVA) [7,24]. In this article, we constructed a predictive model, Neu_LR, to correctly identify presynaptic and postsynaptic neurotoxins. The monoMonokGap method was used to extract the frequency characteristics of presynaptic and postsynaptic neurotoxin sequences and carry out feature selection. Then, based on the important features obtained after dimensionality reduction, the logistic regression algorithm is used to construct the prediction model Neu_LR.

    As the effectiveness of machine learning technology has been continuously verified in recent years, the prediction of protein classification using machine learning-related technology has become a new research category [25,26]. The key to protein classification prediction with the help of machine learning technology lies in data processing and classification algorithms. The general prediction process is to first use the algorithm to extract the features of the protein and then use different classifiers to predict the protein. Therefore, the effective combination of feature extraction algorithms and classifiers has been extensively studied [27,28,29,30,31].

    The research contents of this study include. We first download the presynaptic and postsynaptic neurotoxins from the UniProt database, and then the monoMonokGap feature expression algorithm is used to extract and select features from the data set to get the optimal features. Second, the feature vector obtained by dimensionality reduction was taken as the input, and the model was built using a logistic regression algorithm, and ten-fold cross-validation and independent test set validation was carried out. Figure 1 shows the flow chart of building the model in this paper [32,33].

    Figure 1.  Main flowchart for predicting presynaptic and postsynaptic neurotoxin proteins.

    High-quality data sets are the basis for building reliable and accurate models [34,35]. The UniProt database provides the scientific community with a single, centralized, authoritative source of protein sequence and functional information [36]. The data set used in this article is also applied to the research of Hua Tang et al. A total of 91 presynaptic and 165 postsynaptic neurotoxins were downloaded from the UniProt database. Since fuzzy information will reduce the quality of the benchmark data set and will cause the predicted model to become unreliable, we must eliminate unknown residues in the protein sequence (such as "X", "Z", "J", "O" and "B"). Because of the highly similar protein sequences in the data set, the results can be overestimated; therefore, the cut-off value of sequence identity is set to 80%. According to the results of the above screening, our data set contains 90 presynaptic neurotoxins, 165 postsynaptic neurotoxins and a total of 255 types of neurotoxin samples can be expressed by the following formula:

    S=SprSpo

    where subset Spr is a collection of 90 presynaptic neurotoxins and Spo is a collection of 165 postsynaptic neurotoxins.

    Each protein sequence can be expressed by the following formula:

    R=r1r2r3rL

    R stands for protein sequence, ri stands for representative residue, and L is the length of the protein sequence. Since some machine learning methods cannot directly learn R, the protein sequences have to be converted to fixed-length vectors [37].

    As the first step in building a biological sequence analysis model, feature extraction is an important part of the correct prediction of protein sequences. Generally, we can use the feature extraction method to convert the input neurotoxin sample order into a fixed-length digital vector and then use MRMD2.0 to reduce the dimensionality of the obtained feature vector as needed. Finally, we use the reduced dimensionality vector as the input vector of the classifier model and classify and process the result [38].

    The monoMonoKGap is a feature extraction method, and the best features can be generated from a large number of previously generated features, monoMonoKGap considers kGap in the nucleotide sub-sequence, frequencies of these sub-sequences are treated as prediction features. It can be used for feature extraction of DNA sequences, RNA sequences, and protein sequences [39]. The selection range of kGap in DNA sequences and RNA sequences is 1 to 5, and the selection range of kGap in protein sequences is 1 to 10. When the kGap value is small, the formation of the feature set is small, and the frequency of occurrence of these features retains the partial or short information sequence order, and when the kGap value is moderately large, more features will be generated to retain long sequence information [40]. Specifically, when kGap = 1, the sequence can be encoded as frequencies of X_X, Simultaneously generate 4×1×4 dimensional features, or 20×1×20 dimensional features, when kGap = 2, generate 4×2×4 dimensional features, or 20×2×20 dimensional features and so on [41]. That is when kGap = n, the DNA sequence and RNA sequence will produce 4×4×n features, and the protein sequence will produce 20×20×n features. The generated feature format is as follows:

    when kGap = 1, the characteristic structure is X_X;

    when kGap = 2, the characteristic structure is X_X and X__X; and so on.

    where X is defined as:

    X={{A,C,G,T},{A,C,G,U}{A,C,D,E,F,G,H,I,K,L,M,N,P,Q,R,S,T,V,W,Y}

    This is followed by the DNA sequence, RNA sequence, and protein sequence.

    For feature selection, to reduce the negative impact of dimensionality and to maintain information features, the AdaBoost classification model can be used to calculate the average impurity reduction. AdaBoost is a popular boosting classification algorithm in data mining. The core idea is to train multiple weak classifiers on the same training set, set these weak classifiers, and finally build a strong classifier [42]. Since Freund and Schapire proposed the AdaBoost algorithm [43,44,45], the improvement of AdaBoost has mainly involved two aspects: 1) adjusting the weight of the weak classifier in a new way and 2) improving the training method to reduce the error rate of classifier or save the training time.

    Each residue in the protein sequence has many physical and chemical properties, so the protein sequence can be regarded as a time series with corresponding properties [46]. PSI-BIAST is run to compare the data set and parameters to generate the outline of each sequence.

    AC-PSSM [47] uses PSI-BIAST [48] as a search tool. PSI-BIAST provides a means to detect distant relationships between proteins. It is a protein sequence profile search method used to compare more sensitive protein sequences to other protein sequences. The database used is the nr library.

    AC-PSSM can convert PSSMs of different lengths into vectors of fixed length. AC measures the correlation between two residues with the same property, which can be expressed as:

    {AC(i,lag)=Llagj=1(Si,jˉSi)(Si,j+lagˉSi)/(Llag)ˉSi=Lj=1Si,j/L (1)

    where i is one of the residues, L is the length of the protein sequence, Si,j is the PSSM fraction of amino acid i at the jth position, and ˉSi is the average fraction of amino acid i in the entire protein sequence. In this way, the number of ACs can be calculated by 20×LAG, where LAG is the maximum hysteresis, where the value of lag is all integers from 1 to LAG. In this article, we set the value of LAG to the default value of 2, that is, the maximum hysteresis is 2. Figure 2 shows the flowchart for generating AC-PSSM when the hysteresis is 2. That is, when LAG = 2, AC-PSSM can generate 40 columns of feature vectors. The first 20 columns represent the characteristics of the lagging item in the PSSM matrix when lag = 1, expressed as

    (A,A,1),(R,R,1),(N,N,1),,(V,V,1)
    Figure 2.  A flowchart of AC-PSSM generation. When LAG = 2, a 40-column feature vector is generated. (A) First generate the first 20 columns, that is, the features when one item is lagging. (B) Secondly, the characteristics of the last 20 columns that are lagging two items are generated.

    The last 20 columns represent the characteristics of the two lags in the PSSM matrix when lag = 2, expressed as

    (A,A,2),(R,R,2),(N,N,2),,(V,V,2)

    Feature selection is also known as variable selection or attribute selection, defined as the process of selecting the feature that contributes the most to the predictor variable or output of interest. After extracting features from the sequence, the MRMD2.0 algorithm is used for feature selection.

    MRMD2.0 is based on the PageRank algorithm. It is not only a Python-based tool for reducing the dimensionality of data sets but also draws performance curves based on feature dimensions. Accuracy can be sacrificed for fewer features by selecting dimensions from the performance curve [49,50].

    The basic idea of classification is to learn the parameters of the classifier through training data, and the goal of classification is to train the parameters of the classifier with the training set with the smallest loss of accuracy [51,52]. Weka (3.8.5) can be used for data mining and prediction models. There are many different classification modes, such as random forest and Bayesian classifiers [27,49,53].

    The random forest algorithm is an ensemble algorithm, which is composed of multiple decision tree classifiers, and each subclassifier is a CART classification regression tree; therefore, classification and regression are performed using random forest. This algorithm is highly resistant to overfitting: the risk of overfitting can be reduced by averaging the decision tree [54,55]. The advantages are its simple implementation, high accuracy, fast training speed, strong anti-overfitting ability, and suitability as a benchmark model. The disadvantage is that the model is prone to overfitting on some sample sets with relatively large noise. When there are many decision trees, the training time and space will be relatively large.

    Logistic regression is a machine learning method used to solve binary classification problems, and it is a generalized linear regression [56]. A hyperplane can be established to classify samples, which can be described by the following formula:

    hθ(x)=g(θTx)=11+eθTx (2)

    where X is the sample x=[x1,x2,,xn] is a n-dimensional vector, g is a logistic function, and the general form is g(z)=11+ez. The advantages are that the computational cost is not high and it is easy to understand and implement, but the disadvantages are that it is easy to underfit and the classification accuracy may not be high.

    In this article, k-fold cross-validation was chosen to test predictions. Specificity (SP), sensitivity (SN), and accuracy (ACC) [57,58,59,60,61] were used to evaluate our proposed method [62]; they can be expressed as:

    SN=1NprpoNpr (3)
    SP=1NpoprNpo (4)
    ACC=1Nprpo+NpoprNpr+Npo (5)

    where Npr and Npo represent presynaptic neurotoxin and postsynaptic neurotoxin, respectively; Nprpo indicates that the presynaptic neurotoxin was incorrectly predicted as a postsynaptic neurotoxin; and Npopr indicates that a postsynaptic neurotoxin was incorrectly predicted as a presynaptic neurotoxin.

    Previous studies have shown that feature extraction is very important for predictor variables, and optimized features can improve the accuracy of model prediction [63,64,65,66,67]. Especially in some high-dimensional data, there may be some noise and redundant information, which will have some negative effects on the prediction.

    In this section, the prediction results of monoMonokGap, AC-PSSM, 188D characteristics, Geary-related characteristics, and amino acid composition (AAC) [68,69] characteristics under logistic regression and random forest were compared. The results are shown in Tables 1 and 2 (where the maximum value is indicated in bold). It can be seen from Tables 1 and 2 that the monoMonokGap feature used in this model has the best performance in all indicators. Under the random forest algorithm, when kGap = 7, monoMonokGap (kGap = 7) performs the best; Under the logistic regression algorithm, when kGap = 9, monoMonokGap (kGap = 9) achieves the best result. Compared with the two results, kGap = 9 had the best performance on LR, among which the values of ACC, AUC, SP, and SN were 99.6078%, 0.996, 0.998, and 0.996, respectively. This also proves the effectiveness of monoMonokGap (kGap = 9), so we choose monoMonokGap (kGap = 9) as the feature expression method of the model.

    Table 1.  Different feature representation methods are used to obtain the results of using RF classifier under 5-fold cross-validation.
    Method SN SP AUC ACC (%)
    monoMonokGap (kGap=1) 0.910 0.870 0.966 90.9804
    monoMonokGap (kGap=3) 0.941 0.912 0.985 94.1176
    monoMonokGap (kGap=5) 0.945 0.920 0.992 94.5098
    monoMonokGap (kGap=7) 0.957 0.936 0.991 95.6863
    monoMonokGap (kGap=9) 0.949 0.917 0.992 94.902
    AC-PSSM 0.839 0.776 0.868 83.9216
    188D 0.710 0.609 0.794 70.9804
    Geary 0.533 0.417 0.481 53.3333
    AAC 0.686 0.602 0.736 68.6275

     | Show Table
    DownLoad: CSV
    Table 2.  Different feature representation methods are used to obtain the results of using LR classifier under 10-fold cross-validation.
    Method SN SP AUC ACC (%)
    monoMonokGap (kGap=1) 0.929 0.901 0.971 92.9412
    monoMonokGap (kGap=3) 0.925 0.889 0.988 92.549
    monoMonokGap (kGap=5) 0.984 0.986 0.994 98.4314
    monoMonokGap (kGap=7) 0.973 0.980 0.995 97.2549
    monoMonokGap (kGap=9) 0.996 0.998 0.996 99.6078
    AC-PSSM 0.780 0.663 0.789 78.040
    188D 0.631 0.37 0.603 63.1373
    Geary 0.627 0.428 0.684 62.7451
    AAC 0.667 0.460 0.637 66.6667

     | Show Table
    DownLoad: CSV

    To further verify the stability of monoMonokGap, we tested it with an independent test set [70] and compared it with AC-PSSM, 188D characteristics, Geary-related characteristics, AAC and other feature expression methods. The results are shown in Tables 3 and 4. Among them, 80% of the randomly selected data sets are used to train the prediction model, and the remaining 20% are used to test the model.

    Table 3.  The performance comparison of different feature algorithms under LR was verified by an independent test set.
    Method SN SP AUC ACC (%)
    monoMonokGap (kGap=1) 0.824 0.828 0.931 82.3592
    monoMonokGap (kGap=3) 0.882 0.835 0.894 88.22353
    monoMonokGap (kGap=5) 0.980 0.989 0.985 98.0392
    monoMonokGap (kGap=7) 0.961 0.979 0.998 96.0784
    monoMonokGap (kGap=9) 0.941 0.968 0.977 94.1176
    AC-PSSM 0.686 0.677 0.709 68.6275
    188D 0.373 0.658 0.608 37.2549
    Geary 0.588 0.397 0.396 58.8235
    AAC 0.604 0.445 0.548 60.3774

     | Show Table
    DownLoad: CSV
    Table 4.  The performance comparison of different feature algorithms under RF was verified by an independent test set.
    Method SN SP AUC ACC (%)
    monoMonokGap (kGap=1) 0.804 0.742 0.872 80.3922
    monoMonokGap (kGap=3) 0.804 0.716 0.923 80.3922
    monoMonokGap (kGap=5) 0.902 0.846 0.966 90.1961
    monoMonokGap (kGap=7) 0.882 0.810 0.977 88.2353
    monoMonokGap (kGap=9) 0.922 0.881 0.973 92.1569
    AC-PSSM 0.725 0.598 0.697 72.549
    188D 0.510 0.707 0.786 50.9804
    Geary 0.529 0.390 0.412 52.9412
    AAC 0.566 0.501 0.553 56.6038

     | Show Table
    DownLoad: CSV

    It can be seen from Tables 3 and 4 that, compared with other feature expression methods, the monoMonokGap feature expression method selected in this article has little difference in the results of independent verification set to test and the results of ten-fold cross-validation, which also proves that the feature expression method selected in this paper does not have over-fitting and has good stability.

    In this section, the performance of the model constructed in this article is compared with RF, Logical Model Tree (LMT), J48, Bayesnet, NaiveBayes, Sequential Minimal Optimization (SMO), and other classifiers. The results of ten-fold cross-validation are shown in Table 5 (the maximum value is indicated in bold). It can be seen from Table 5 that the results of the model constructed in this paper are significantly better than those of other classifiers in terms of various indicators, with the values of ACC, AUC, SP, and SN being 99.6078%, 0.996, 0.998 and 0.996 respectively. This also proves the validity of the model constructed in this article.

    Table 5.  Comparison of performance of different classification algorithms under 10-fold cross-validation.
    Classifier SN SP AUC ACC (%)
    Neu_LR 0.996 0.998 0.996 99.6078
    RF 0969 0.948 0.995 96.8627
    LMT 0.918 0.884 0.971 91.7647
    J48 0.863 0.804 0.877 86.2745
    BayesNet 0.929 0.891 0.987 92.9412
    NaiveBayes 0.863 0.779 0.945 86.2745
    SMO 0.973 0.965 0.969 97.2549

     | Show Table
    DownLoad: CSV

    To further verify the robustness of the model constructed in this article, we conducted independent test set verification on it and compared its performance with RF, LMT, J48, Bayesnet, NaiveBayes, SMO, and other classifiers. The results are shown in Table 6. Among them, 80% of the randomly selected data sets are used to train the prediction model, and the remaining 20% are used to test the model.

    Table 6.  The performance comparison of different classification algorithms was verified by an independent test set.
    Classifier SN SP AUC ACC (%)
    Neu_LR 0.941 0.968 0.977 94.1176
    RF 0.922 0.881 0.972 92.1569
    LMT 0.784 0.756 0.886 78.4314
    J48 0.843 0.788 0.824 84.3137
    BayesNet 0.902 0.846 0.948 90.1961
    NaiveBayes 0.804 0.691 0.920 80.3922
    SMO 0.922 0.932 0.927 92.1569

     | Show Table
    DownLoad: CSV

    It can be seen from Table 6 that, compared to other algorithms, the model constructed in this paper achieves the best prediction article, and the values of ACC, AUC, SP, and SN are 94.1176%, 0.977, 0.968, and 0.941, respectively. Moreover, compared with the results of 10-fold cross-validation, the difference is very small, which also proves that the prediction performance of the model constructed in this article does not exist overfitting and has good stability.

    This section compares the model constructed in this paper with other existing methods. The comparison results are shown in Table 7, where the results of ID [23] and ANOVA [7] are obtained directly from literature. It can be seen from Table 7 that the model Neu_LR constructed in this paper has the best performance in all indicators, among which ACC, SP, and SN reach the maximum value of 99.6078%, 0.998, and 0.996 respectively, and the effect is better than the other two methods, which also proves the effectiveness of the model Neu_LR constructed in this paper [71].

    Table 7.  Comparison of different methods for predicting presynaptic and postsynaptic neurotoxins.
    Method SN SP ACC (%)
    ID [23] 88.46 91.30 89.80
    ANOVA [7] 94.51 95.15 94.92
    Neu_LR 0.996 0.998 99.6078

     | Show Table
    DownLoad: CSV

    The correct understanding of presynaptic and postsynaptic neurotoxins is an essential first step in the discovery of drug targets and drug design. And protein prediction mainly involves two aspects, feature extraction, and selection of classification algorithms. Therefore, the prediction model Neu_LR was constructed in this article. The monoMonokGap method was used to extract the frequency characteristics of presynaptic and postsynaptic neurotoxin sequences, and the feature selection was carried out. Then, based on the important features obtained after dimension-reduction, the logistic regression algorithm was used to construct the prediction model Neu_LR. In this paper, we use 10-fold cross-validation and independent test set validation to judge whether the Neu_LR model is good or not. In the 10-fold cross-validation, we achieved 99.6078% accuracy, and in the independent test set validation, we achieved 94.1176% accuracy, which shows that our model is feasible and effective.

    This work was supported by the National Nature Science Foundation of China (Grant Nos 61863010, 11926205, 11926412, and 61873076), National Key R & D Program of China (No.2020YFB2104400) and Natural Science Foundation of Hainan, China (Grant Nos. 119MS036 and 120RC588), and Hainan Normal University 2020 Graduate Student Innovation Research Project (hsyx2020-41), the Special Science Foundation of Quzhou (2020D003).

    All authors declare no conflicts of interest in this paper.



    [1] Abaje IB, Bello Y, Ahmad SA (2020) A review of air quality and concentrations of air pollutants in Nigeria. JASEM 24: 373–379. https://doi.org/10.4314/jasem.v24i2.25 doi: 10.4314/jasem.v24i2.25
    [2] Schroeter LC (1966) Sulfur dioxide: Applications in foods, beverages, and pharmaceuticals, Long Island City: Pergamon Press, Inc.
    [3] Mandrile L, Cagnasso I, Berta L, et al. (2020) Direct quantification of sulfur dioxide in wine by Surface Enhanced Raman Spectroscopy. Food Chem 326: 127009. https://doi.org/10.1016/j.foodchem.2020.127009 doi: 10.1016/j.foodchem.2020.127009
    [4] Morgan SC, Tantikachornkiat M, Scholl CM, et al. (2019) The effect of sulfur dioxide addition at crush on the fungal and bacterial communities and the sensory attributes of Pinot gris wines. Int J Food Microbiol 290: 1–14. https://doi.org/10.1016/j.ijfoodmicro.2018.09.020 doi: 10.1016/j.ijfoodmicro.2018.09.020
    [5] Gabriele M, Gerardi C, Lucejko JJ, et al. (2018) Effects of low sulfur dioxide concentrations on bioactive compounds and antioxidant properties of Aglianico red wine. Food Chem 245: 1105–1112. https://doi.org/10.1016/j.foodchem.2017.11.060 doi: 10.1016/j.foodchem.2017.11.060
    [6] Giacosa S, Segade SR, Cagnasso E, et al. (2019) Red Wine Technology, 1 Eds., New York: Academic Press, 309–321. https://doi.org/10.1016/B978-0-12-814399-5.00021-9
    [7] Seralini G, Douzelet J, Halley J (2021) Sulfur in Wines and Vineyards: Taste and Comparative Toxicity to Pesticides. Food Nutr J 6: 231. https://doi.org/10.29011/2575-7091.100131 doi: 10.29011/2575-7091.100131
    [8] Directive 2003/89/EC of the European Parliament and of the Council of 10 November 2003 amending Directive 2000/13/EC as regards indication of the ingredients present in foodstuffs (Text with EEA relevance). (2003) Available from: https://eur-lex.europa.eu/legal-content/en/ALL/?uri=CELEX%3A32003L008913/12/2014.
    [9] Commission Regulation No 606/2009 of 10 July 2009. Laying down certain detailed rules for implementing Council Regulation (EC) No 479/2008 as regards the categories of grapevine products, oenological practices and the applicable restrictions. (2009) Available from: https://eur-lex.europa.eu/eli/reg/2009/606/oj/eng.
    [10] Jenkins TW, Howe PA, Sacks GL, et al. (2020) Determination of molecular and " Truly" free sulfur dioxide in wine: a comparison of headspace and conventional methods. AJEV 71: 222–230. https://doi.org/10.5344/ajev.2020.19052 doi: 10.5344/ajev.2020.19052
    [11] Danchana K, Clavijo S, Cerdà V (2019) Conductometric determination of sulfur dioxide in wine using a multipumping system coupled to a gas-diffusion cell. Anal Lett 52: 1363–1378. https://doi.org/10.1080/00032719.2018.1539742 doi: 10.1080/00032719.2018.1539742
    [12] Eurachem Guide (2014) Eurachem Guide: The Fitness for Purpose of Analytical Methods—A Laboratory Guide to Method Validation and Related Topics, 2 Eds., Available from: https://www.eurachem.org/images/stories/Guides/pdf/MV_guide_2nd_ed_EN.pdf.
    [13] Pepe P, Bosco A, Capuano F, et al. (2021) Towards an Integrated Approach for Monitoring Toxoplasmosis in Southern Italy. Animals 1: 1949. https://doi.org/10.3390/ani11071949 doi: 10.3390/ani11071949
    [14] Laneri S, Di Lorenzo R, Sacchi A, et al. (2019) Dosage of bioactive molecules in the nutricosmeceutical helix aspersa muller mucus and formulation of new cosmetic cream with moisturizing effect. Nat Prod Commun 14: 1–7. https://doi.org/10.1177/1934578X19868606 doi: 10.1177/1934578X19868606
    [15] Mancusi F, Capuano S, Girardi O, et al. (2022) Detection of SARS-CoV-2 RNA in bivalve mollusks by droplet digital RT-PCR (dd RT-PCR). Int J Environ Res Public Health 19: 943. https://doi.org/10.3390/ijerph19020943 doi: 10.3390/ijerph19020943
    [16] Fanelli F, Cozzi G, Raiola A, et al. (2017) Raisins and currants as conventional nutraceuticals in Italian market: Natural occurrence of Ochratoxin A. J Food Sci 82: 2306–2312. https://doi: 10.1111/1750-3841.13854 doi: 10.1111/1750-3841.13854
    [17] Mancusi A, Giordano A, Bosco A, et al. (2022) Development of a droplet digital polymerase chain reaction tool for the detection of Toxoplasma gondii in meat samples. Parasitol Res 121: 1467–1473. https://doi.org/10.1007/s00436-022-07477-9 doi: 10.1007/s00436-022-07477-9
    [18] Capuano F, Capparelli R, Mancusi A, et al. (2013) Detection of Brucella spp. in Stretched Curd Cheese as Assessed by Molecular Assays. J Food Saf 33: 145–148. https://doi.org/10.1111/jfs.12034 doi: 10.1111/jfs.12034
    [19] Cristiano D, Peruzy MF, Aponte M, et al. (2021) Comparison of droplet digital PCR vs real-time PCR for Yersinia enterocolitica detection in vegetables. Int J Food Microbiol 354: 109321. https://doi.org/10.1016/j.ijfoodmicro.2021.109321 doi: 10.1016/j.ijfoodmicro.2021.109321
    [20] OIV 2007 Organisation Internationale de la Vigne et du Vin (2007) Available from: https://www.oiv.int/en/technical-standards-and-documents/methods-of-analysis/compendium-of-international-methods-of-analysis-of-wines-and-musts.
    [21] Shapiro SS, Wilk MB (1965) An analysis of variance test for normality (complete samples). Biometrika 52: 591–611. https://doi.org/10.2307/2333709 doi: 10.2307/2333709
    [22] Huber PJ (1981) Robust statistics, New York: Wiley, 1–384.
    [23] Dini I, Seccia S, Senatore A, et al. (2019) Development and validation of an analytical method for total polyphenols quantification in extra virgin olive oils. Food Anal Met 13: 457–464. https://doi.org/10.1007/s12161-019-01657-7 doi: 10.1007/s12161-019-01657-7
    [24] Dini I, Di Lorenzo R, Senatore A, et al. (2020) Validation of rapid enzymatic quantification of acetic acid in vinegar on automated spectrophotometric system. Foods 9: 761. https://doi.org/10.3390/foods9060761 doi: 10.3390/foods9060761
    [25] ISO/IEC 17025: 2017 General requirements for the competence of testing and calibration laboratories (2017) Available from: https://www.iso.org/standard/66912.html.
    [26] Molognoni L, Daguer H, Dos Santos IR, et al. (2019) Influence of method validation parameters in the measurement uncertainty estimation by experimental approaches in food preservatives analysis. Food Chem 282: 147–152. https://doi.org/10.1016/j.foodchem.2018.12.115 doi: 10.1016/j.foodchem.2018.12.115
    [27] Dini I, Di Lorenzo R, Senatore A, et al. (2021) Comparison between Mid-Infrared (ATR-FTIR) spectroscopy and official analysis methods for determination of the concentrations of alcohol, SO2, and total acids in wine. Separations 8: 191. https://doi.org/10.3390/separations8100191 doi: 10.3390/separations8100191
  • This article has been cited by:

    1. Hao Wan, Qing Liu, Ying Ju, Utilize a few features to classify presynaptic and postsynaptic neurotoxins, 2023, 152, 00104825, 106380, 10.1016/j.compbiomed.2022.106380
    2. Muhammad Nabeel Asim, Muhammad Ali Ibrahim, Ahtisham Fazeel, Andreas Dengel, Sheraz Ahmed, DNA-MP: a generalized DNA modifications predictor for multiple species based on powerful sequence encoding method, 2023, 24, 1467-5463, 10.1093/bib/bbac546
    3. Hongwei Zhang, Yan Shi, Yapeng Wang, Xu Yang, Kefeng Li, Sio-Kei Im, Yu Han, LMFE: A Novel Method for Predicting Plant LncRNA Based on Multi-Feature Fusion and Ensemble Learning, 2025, 16, 2073-4425, 424, 10.3390/genes16040424
  • Reader Comments
  • © 2023 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Metrics

Article views(2202) PDF downloads(278) Cited by(2)

Figures and Tables

Figures(2)  /  Tables(18)

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return

Catalog