Final Support Vector Machine Model Research Articles

Improved prediction of drug-induced liver injury literature using natural language processing and machine learning methods.

Drug-induced liver injury (DILI) is an adverse hepatic drug reaction that can potentially lead to life-threatening liver failure. Previously published work in the scientific literature on DILI has provided valuable insights for the understanding of hepatotoxicity as well as drug development. However, the manual search of scientific literature in PubMed is laborious and time-consuming. Natural language processing (NLP) techniques along with artificial intelligence/machine learning approaches may allow for automatic processing in identifying DILI-related literature, but useful methods are yet to be demonstrated. To address this issue, we have developed an integrated NLP/machine learning classification model to identify DILI-related literature using only paper titles and abstracts. For prediction modeling, we used 14,203 publications provided by the Critical Assessment of Massive Data Analysis (CAMDA) challenge, employing word vectorization techniques in NLP in conjunction with machine learning methods. Classification modeling was performed using 2/3 of the data for training and the remainder for test in internal validation. The best performance was achieved using a linear support vector machine (SVM) model on the combined vectors derived from term frequency-inverse document frequency (TF-IDF) and Word2Vec, resulting in an accuracy of 95.0% and an F1-score of 95.0%. The final SVM model constructed from all 14,203 publications was tested on independent datasets, resulting in accuracies of 92.5%, 96.3%, and 98.3%, and F1-scores of 93.5%, 86.1%, and 75.6% for three test sets (T1-T3). Furthermore, the SVM model was tested on four external validation sets (V1-V4), resulting in accuracies of 92.0%, 96.2%, 98.3%, and 93.1%, and F1-scores of 92.4%, 82.9%, 75.0%, and 93.3%.

Examining the Use of Urban Growth Boundary for Future Urban Expansion of Chattogram, Bangladesh

With the rapid and unregulated nature of urban expansion occurring in Chattogram, Bangladesh, the adoption of urban growth restriction mechanisms such as the urban growth boundary (UGB) can provide a robust framework necessary to direct the development of built-up areas in a way that curtails the growth in environmentally sensitive areas of the city. Using a support vector machine (SVM)-based urban growth simulation model, this paper examines the areas of future contiguous expansion of the city to aid in the delineation of the UGB. Utilizing landcover, topographic, and population density data from a variety of sources for the past twenty years, the SVM method with the radial basis function (RBF) kernel is used to develop a model based on fourteen predictor variables. A grid-search is used to tune the hyperparameters and determine the best performance combination of the hyperparameters for the RBF kernel function used in the SVM. The final SVM model using the best performance combination of the hyperparameters indicates a high percentage agreement of 91.79% and a substantial agreement for the Kappa coefficient of 0.7699. The developed SVM simulation model identifies potential areas that are more likely to undergo urban expansion in Chattogram in the next twenty years and provides aids for a stringent and strict delineation of UGB for this region.

Three-dimensional CT texture analysis of anatomic liver segments can differentiate between low-grade and high-grade fibrosis

BackgroundCT texture analysis (CTTA) has been successfully used to assess tissue heterogeneity in multiple diseases. The purpose of this work is to demonstrate the value of three-dimensional CTTA in the evaluation of diffuse liver disease. We aimed to develop CTTA based prediction models, which can be used for staging of fibrosis in different anatomic liver segments irrespective of variations in scanning parameters.MethodsWe retrospectively collected CT scans of thirty-two chronic hepatitis patients with liver fibrosis. The CT examinations were performed on either a 16- or a 64-slice scanner. Altogether 354 anatomic liver segments were manually highlighted on portal venous phase images, and 1117 three-dimensional texture parameters were calculated from each segment. The segments were divided between groups of low-grade and high-grade fibrosis using shear-wave elastography. The highly-correlated features (Pearson r > 0.95) were filtered out, and the remaining 453 features were normalized and used in a classification with k-means and hierarchical cluster analysis. The segments were split between the train and test sets in equal proportion (analysis I) or based on the scanner type (analysis II) into 64-slice train 16-slice validation cohorts for machine learning classification, and a subset of highly prognostic features was selected with recursive feature elimination.ResultsA classification with k-means and hierarchical cluster analysis divided segments into four main clusters. The average CT density was significantly higher in cluster-4 (110 HU ± SD = 10.1HU) compared to the other clusters (c1: 96.1 HU ± SD = 11.3HU; p < 0.0001; c2: 90.8 HU ± SD = 16.8HU; p < 0.0001; c3: 93.1 HU ± SD = 17.5HU; p < 0.0001); but there was no difference in liver stiffness or scanner type among the clusters. The optimized random forest classifier was able to distinguish between low-grade and high-grade fibrosis with excellent cross-validated accuracy in both the first and second analysis (AUC = 0.90, CI = 0.85–0.95 vs. AUC = 0.88, CI = 0.84–0.91). The final support vector machine model achieved an excellent prediction rate in the second analysis (AUC = 0.91, CI = 0.88–0.94) and an acceptable prediction rate in the first analysis (AUC = 0.76, CI = 0.67–0.84).ConclusionsIn conclusion, CTTA-based models can be successfully applied to differentiate high-grade from low-grade fibrosis irrespective of the imaging platform. Thus, CTTA may be useful in the non-invasive prognostication of patients with chronic liver disease.

Severity Detection for the Coronavirus Disease 2019 (COVID-19) Patients Using a Machine Learning Model Based on the Blood and Urine Tests.

The recent outbreak of the coronavirus disease-2019 (COVID-19) caused serious challenges to the human society in China and across the world. COVID-19 induced pneumonia in human hosts and carried a highly inter-person contagiousness. The COVID-19 patients may carry severe symptoms, and some of them may even die of major organ failures. This study utilized the machine learning algorithms to build the COVID-19 severeness detection model. Support vector machine (SVM) demonstrated a promising detection accuracy after 32 features were detected to be significantly associated with the COVID-19 severeness. These 32 features were further screened for inter-feature redundancies. The final SVM model was trained using 28 features and achieved the overall accuracy 0.8148. This work may facilitate the risk estimation of whether the COVID-19 patients would develop the severe symptoms. The 28 COVID-19 severeness associated biomarkers may also be investigated for their underlining mechanisms how they were involved in the COVID-19 infections.

Severity Detection&amp;nbsp;For the Coronavirus Disease 2019 (COVID-19) Patients Using a Machine Learning Model Based on the Blood and Urine Tests

Background: The recent outbreak of the coronavirus disease-2019 (COVID-19) caused serious challenges to the human society in China and across the world. COVID-19 induced pneumonia in human hosts and carried a highly inter-person contagiousness. The COVID-19 patients may carry severe symptoms, and some of them may even die of major organ failures. We aim to utilize the machine learning algorithms to build the COVID-19 severeness detection model. Methods: This study recruited the binary classification problem between 75 severely illed COVID-19 infected patients and the other 62 patients with mild symptoms. Support vector machine (SVM) demonstrated a promising detection accuracy after 32 features were detected to be significantly associated with the COVID-19 severeness. These 32 features were further screened for inter-feature redundancies. Findings: The severely illed patients had a higher serum level of neutrophil percentage and lower serum levels of monocyte percentage and calcium compared with those mild ones. The blood test features demonstrated much more significant inter-group differences than the urine test features. These three blood test features as candidate severeness biomarkers, i.e., serum ferritin, hs-CRP, interleukin-2R, and tumor necrosis factor-α. The final SVM model achieved the overall accuracy 0.8148 using 28 features. Interpretation: This study utilized the machine learning algorithms to detect the COVID-19 severely ill patients from those with only mild symptoms. Our experimental data demonstrated strong correlations with the COVID-19 severeness. The 28 biomarkers may also be investigated for their underlining mechanisms of their roles in the COVID-19 severely ill patients. Funding Statement: This work was supported by grants from The epidemiology, early warning and response techniques of major infectious diseases in the Belt and Road Initiative (#2018ZX10101002), National Natural Science Foundation of China (#81871699), Jilin Provincial Key Laboratory of Big Data Intelligent Computing (20180622002JC), the Education Department of Jilin Province (JJKH20180145KJ), Foundation of Jilin Province Science and Technology Department (#172408GH010234983), and the startup grant of the Jilin University. This work was also partially supported by the Bioknow MedAI Institute (BMCPP-2018-001), the High Performance Computing Center of Jilin University, and the Fundamental Research Funds for the Central Universities, JLU. Declaration of Interests: The authors declare no competing interests. Ethics Approval Statement: This study was approved by the Ethics Commission of the First Hospital of Jilin University. Informed consent was waived for this emerging infectious disease.

Prediction of human clearance based on animal data and molecular properties.

Human clearance is often predicted prior to clinical study from invivo preclinical data by virtue of interspecies allometric scaling methods. The aims of this study were to determine the important molecular descriptors for the extrapolation of animal data to human clearance and further to build a model to predict human clearance by combination of animal data and the selected molecular descriptors. These important molecular descriptors selected by genetic algorithm (GA) were from five classes: quantum mechanical, shadow indices, E-state keys, molecular properties, and molecular property counts. Although the data set contained many outliers determined by the conventional Mahmood method, the variation of most outliers was reduced significantly by our final support vector machine (SVM) model. The values of cross-validated correlation coefficient and root-mean-squared error (RMSE) for leave-one-out cross-validation (LOOCV) of the final SVM model were 0.783 and 0.305, respectively. Meanwhile, the reliability and consistency of the final model were also validated by an external test set. In conclusion, the SVM model based on the molecular descriptors selected by GA and animal data achieved better prediction performance than the Mahmood method. This approach can be applied as an improved interspecies allometric scaling method in drug research and development.

Predicting substrates of the human breast cancer resistance protein using a support vector machine method

BackgroundHuman breast cancer resistance protein (BCRP) is an ATP-binding cassette (ABC) efflux transporter that confers multidrug resistance in cancers and also plays an important role in the absorption, distribution and elimination of drugs. Prediction as to if drugs or new molecular entities are BCRP substrates should afford a cost-effective means that can help evaluate the pharmacokinetic properties, efficacy, and safety of these drugs or drug candidates. At present, limited studies have been done to develop in silico prediction models for BCRP substrates.In this study, we developed support vector machine (SVM) models to predict wild-type BCRP substrates based on a total of 263 known BCRP substrates and non-substrates collected from literature. The final SVM model was integrated to a free web server.ResultsWe showed that the final SVM model had an overall prediction accuracy of ~73% for an independent external validation data set of 40 compounds. The prediction accuracy for wild-type BCRP substrates was ~76%, which is higher than that for non-substrates. The free web server (http://bcrp.althotas.com) allows the users to predict whether a query compound is a wild-type BCRP substrate and calculate its physicochemical properties such as molecular weight, logP value, and polarizability.ConclusionsWe have developed an SVM prediction model for wild-type BCRP substrates based on a relatively large number of known wild-type BCRP substrates and non-substrates. This model may prove valuable for screening substrates and non-substrates of BCRP, a clinically important ABC efflux drug transporter.

The prediction of promoter sequences based on the chemical features

A novel strategy to the prediction of promoter sequences, which was based on the analysis of chemical features, was developed for the first time. A string of nucleotides was translated into numerical sequences by means of the chemical parameters that represented the chemical properties and molecular structures of nucleotides, and then genetic algorithm was employed to select effective chemical features so as to establish the proper predictive model of support vector machine (SVM). The accuracies of the final SVM model for the leave-one-out cross-validation on the training set were 100%, and the sensitivity and specificity reached to 1. The accuracy for testing set was also up to 100%. Moreover, several functional sites and chemical parameters selected by SVM model were discussed. The satisfactory results indicated that the study of chemical features in sequences was effective, and the proposed approach was reliable.

Prediction of chemical carcinogenicity by machine learning approaches

In this paper we report a successful application of machine learning approaches to the prediction of chemical carcinogenicity. Two different approaches, namely a support vector machine (SVM) and artificial neural network (ANN), were evaluated for predicting chemical carcinogenicity from molecular structure descriptors. A diverse set of 844 compounds, including 600 carcinogenic (CG+) and 244 noncarcinogenic (CG−) molecules, was used to estimate the accuracies of these approaches. The database was divided into two sets: the model construction set and the independent test set. Relevant molecular descriptors were selected by a hybrid feature selection method combining Fischer's score and Monte Carlo simulated annealing from a wide set of molecular descriptors, including physiochemical properties, constitutional, topological, and geometrical descriptors. The first model validation method was based a five-fold cross-validation method, in which the model construction set is split into five subsets. The five-fold cross-validation was used to select descriptors and optimise the model parameters by maximising the averaged overall accuracy. The final SVM model gave an averaged prediction accuracy of 90.7% for CG+ compounds, 81.6% for CG− compounds and 88.1% for the overall accuracy, while the corresponding ANN model provided an averaged prediction accuracy of 86.1% for CG+ compounds, 79.3% for CG− compounds and 84.2% for the overall accuracy. These results indicate that the hybrid feature selection method is very efficient and the selected descriptors are truly relevant to the carcinogenicity of compounds. Another model validation method, i.e. a hold-out method, was used to build the classification model using the selected descriptors and the optimised model parameters, in which the whole model construction set was used to build the classification model and the independent test set was used to test the predictive ability of the model. The SVM model gave a prediction accuracy of 87.6% for CG+ compounds, 79.1% for CG− compounds and 85.0% for the overall accuracy. The ANN model gave a prediction accuracy of 85.6% for CG+ compounds, 79.1% for CG− compounds and 83.6% for the overall accuracy. The results indicate that the built models are potentially useful for facilitating the prediction of chemical carcinogenicity of untested compounds.

Better prediction of the location of α‐turns in proteins with support vector machine

We have developed a novel method named AlphaTurn to predict alpha-turns in proteins based on the support vector machine (SVM). The prediction was done on a data set of 469 nonhomologous proteins containing 967 alpha-turns. A great improvement in prediction performance was achieved by using multiple sequence alignment generated by PSI-BLAST as input instead of the single amino acid sequence. The introduction of secondary structure information predicted by PSIPRED also improved the prediction performance. Moreover, we handled the very uneven data set by combining the cost factor j with the "state-shifting" rule. This further promoted the prediction quality of our method. The final SVM model yielded a Matthews correlation coefficient (MCC) of 0.25 by a 10-fold cross-validation. To our knowledge, this MCC value is the highest obtained so far for predicting alpha-turns. An online Web server based on this method has been developed and can be freely accessed at http://bmc.hust.edu.cn/bioinformatics/ or http://210.42.106.80/.

Prediction of π-turns in proteins using PSI-BLAST profiles and secondary structure information

Due to the structural and functional importance of tight turns, some methods have been proposed to predict γ-turns, β-turns, and α-turns in proteins. In the past, studies of π-turns were made, but not a single prediction approach has been developed so far. It will be useful to develop a method for identifying π-turns in a protein sequence. In this paper, the support vector machine (SVM) method has been introduced to predict π-turns from the amino acid sequence. The training and testing of this approach is performed with a newly collected data set of 640 non-homologous protein chains containing 1931 π-turns. Different sequence encoding schemes have been explored in order to investigate their effects on the prediction performance. With multiple sequence alignment and predicted secondary structure, the final SVM model yields a Matthews correlation coefficient (MCC) of 0.556 by a 7-fold cross-validation. A web server implementing the prediction method is available at the following URL: http://210.42.106.80/piturn/.