Inquiry (E-Journal - Faculty of Business and Administration, International University of Sarajevo)
Not a member yet
209 research outputs found
Sort by
Historical Criteria for Structural Classes of Proteins in Percentages: After 20 Years
Get PDFTwo decades ago scientists proposed some criteria for the structural classes in percentages. Today experts at SCOP classified hundreds of thousands of proteins into one of the four structural classes manually by inspection, and observation. Nakashima et al. gave a classification criteria. P.Y. Chou also proposed another method to classify proteins according their residue contents in three conformations, helix, sheet, and coil. Later P.Y. Chou revised his method. Today SCOP listed around 100.000 proteins with their structural classes. In this paper two datasets will be used to reveal the percentages of residues in α-Helices, β-sheets, and coils in proteins of classes all α, all β, α+β, and α/β, in the classifications made by experts in SCOP. The first of the data bases is PDBselect25 which contains 1670 twilight zone proteins whose similarity is less than 25%. The second data base BF30 consists of 10294 proteins picked from PDB database with the similarity threshold of 30%. Structural classes of these proteins are taken from SCOP database. It is seen that there is a very poor correlation between historical criteria, and SCOP’s scientists’ intuition in classification of proteins into structural classes
COMPARISON OF MACHINE LEARNING TECHNIQUES IN SPAM E-MAIL CLASSIFICATION
Get PDFE-mail still proves to be very popular and an efficient communication tool. Due to its misuse, however, managing e-mails is an important problem for organizations and individuals. Spam, known as unwanted message, is an example of misuse. Specifically, spam is defined as the arrival of unwelcomed bulk email not being requested for by recipients. This paper compares different Machine Learning Techniques in classification of spam e-mails. Random Forest (RF), C4.5 decision tree and Artificial Neural Network (ANN) were tested to determine which method provides the best results in spam e-mail classification. Our results show that RF is the best technique applied on dataset from HP Labs, indicating that ensemble methods may have an edge in spam detectio
Protein Secondary Structure Prediction Based on Physicochemical Features and PSSM by KNN
Get PDFIn this paper, we propose a protein secondary structure prediction method based on the k-nearest neighborhood (KNN) technique with position-specific scoring matrix (PSSM) profiles, propensity matrix of amino acids in three conformations (HEC) and three physicochemical features; hydrophobicity, net charges, and side chain mass. First, the KNN with the optimal k-value is found. Then, the Euclidean distance of 26-dimensional data for each amino acid of a protein, to the data vectors of all other proteins are computed. The conformations of the nearest seven amino acids are pooled. Majority of the pooled votes is given to the amino acid of the quarry protein as the conformation H, E, or C. Finally, we use a filter to refine the predicted results from KNN. After filtering, the accuracy of the prediction goes up to the level of 90% for some proteins. This validates that considering PSSM, the propensity matrix, and physicochemical features may exhibit better performance
A study in Authorship Attribution: The Federalist Papers
Get PDFIn order to authorship attribution techniques, the Federalist Papers have been applied as a testing-ground that twelve of which are claimed by Alexander Hamilton and James Madison. The value of novel stylometric techniques through implementation of them to the Federalist problem is what the paper subjects to. Support vector machines and nearest neighbor techniques alongside Artificial Neural Network techniques are used for classification of selected disputed paper. Encouraging results achieved in the research
Conformational Parameters for Amino Acids in Helical, β-Sheet, and Random Coil Regions Calculated from Proteins: After 40 Years
No full textForty years ago, Peter Y. Chou and Gerald D. Fasman (1974a), relying on the information from fifteen proteins calculated α helix, β-sheet, and coil conformational parameters, Pα,Pβ,and Pc,for the 20 naturally occurring amino acids from the frequency of occurrence of each amino acid residue in the α, β, and coil conformations. Secondary structure of these 15 proteins had been determined by X-ray crystallography. Although the accuracy could not go over to the level of 60% too much, these values utilized for a long time to provide a simple procedure, devoid of complex computer calculations, to predict the secondary structure of proteins from their known amino acid sequences. In the same article of Peter Y. Chou and Gerald D. Fasman, a detailed analysis of the helix and β-sheet boundary residues in proteins provided amino acid frequencies at the N-and C-terminal ends which were used to delineate helical and β regions. Charged residues are found with the greatest frequency at both helical ends, but they were mostly absent in β-sheet regions. In the same article a mechanism of protein folding was proposed, whereby helix nucleation starts at the centers of the helix where the Pα values are highest, and propagates in both directions, until strong helix breakers where Pα values are lowest, terminate the growth at both ends. Similarly, residues with the highest Pβ values will initiate β regions and residues with the lowest Pβ values will terminate β regions. The helical region with the largest Pα was proposed as the site of the first fold during protein renaturation. The mechanism whereby proteins fold into their native conformation, capable of biological activity, has been a long sought after goal. With the elucidation of the three-dimensional structure of many proteins through X-ray crystallography, a new momentum has been given to understanding the factors governing this complex assembly of polypeptide chains. In this paper, using similar statistics from 20 347 proteins, the level of reliability of formerly found results is discussed
Prediction of Protein Structural Classes for Low-Similarity Sequences Based On Predicted Secondary Structure
No full textKnowledge about structural classes of proteins plays an important role in inferring tertiary structure and function of a protein. One of the major problems with the existing algorithm for the prediction of protein structural classes is low accuracies for proteins from + and / classes. To improve accuracies, one needs to extract features with high representation power. Several authors proposed enormous number of features. Some of them redundant, most of them overlapping. In this paper, most prominent features proposed in the literature are reviewed. Features extracted from Position Specific Scoring Matrices (PSSM) are excluded and left as the subject matter of another paper. Also some combinations of these features are used to classify a low-homology dataset, 25PDB, and 30FB, with sequence similarity lower than 25% and 30%, respectively. Comparison of our results with others shows that to find the best combination is very important and may provide a cost-effective alternative to predict protein structural class in particular for low-similarity datasets
The Impact of Mspca Signal De-Noising In Real-Time Wireless Brain Computer Interface System
No full textThis paper presents the practical implementation of the motor imagery BCI system using MATLAB GUI. EEG signals were recorded using Mindwave Mobile Headset from one subject for two motor imagery tasks: right hand and left hand. The offline analysis showed decent performance of the combination between MSPCA de-noising of EEG signals and statistical features extracted from WPD sub-bands. The best classifier from the offline analysis was used in the online assessment to classify new motor imagery EEG signals. The overall results show that the desirable de-noising results are obtained if MSPCA is applied on a data matrix containing signals that belong to one particular class
A Nonparametric Approach to Pricing Options Learning Networks
Get PDFFor practitioners of equity markets, option pricing is a major challenge during high volatility periods and Black-Scholes formula for option pricing is not the proper tool for very deep out-of-the-money options. The Black-Scholes pricing errors are larger in the deeper out-of-the money options relative to the near the-money options, and it's mispricing worsens with increased volatility. Experts opinion is that the Black-Scholes model is not the proper pricing tool in high volatility situations especially for very deep out-of-the-money options. They also argue that prior to the 1987 crash, volatilities were symmetric around zero moneyness, with in-the-money and out-of-the money having higher implied volatilities than at-the-money options. However, after the crash, the call option implied volatilities were decreasing monotonically as the call went deeper into out-of-the-money, while the put option implied volatilities were decreasing monotonically as the put went deeper into in-the-money. Since these findings cannot be explained by the Black-Scholes model and its variations, researchers searched for improved option pricing models. Feedforward networks provide more accurate pricing estimates for the deeper out-of-the money options and handles pricing during high volatility with considerably lower errors for out-of-the-money call and put options. This could be invaluable information for practitioners as option pricing is a major challenge during high volatility periods. In this article a nonparametric method for estimating S&P 100 index option prices using artificial neural networks is presented. To show the value of artificial neural network pricing formulas, Black-Scholes option prices are compared with the network prices against market prices. To illustrate the practical relevance of the network pricing approach, it is applied to the pricing of S&P 100 index options from April 4, 2014 to April 9, 2014. On the five days data while Black-Scholes formula prices have a mean 1.98 for calls, while neural network’s error is less than 1 for calls
A Study on Business Ethics and Corporate Social Responsibility (CSR): Evidence from Bosnia and Herzegovina
Get PDFThis paper tries to explore issues in terms of business ethics and CSR in Bosnia and Herzegovina. We examined 54 companies by using convenience sampling method. We explored descriptive statistics and non probability data analysis. With Spearman’s correlation test we found possible relationships within some issues within the survey parameters that we examined. Conclusions have been derived from this study. For analysis we used a software for social sciences namely SPSS, version 21
Predicting the Secondary Structure of Proteins Using Artificial Neural Networks
Get PDFA method for protein secondary structure prediction based on the use of artificial neural networks (ANN) is presented. Amino acids, and their secondary structures obtained from National Center for Biotechnology Information (NCBI) and the online tool given in Chou-Fasman website of seven proteins are concatenated to create a sequence of 15536 residues. A neural network with only an input and an output layer is used, and back-propagation technique is adopted to tune the synaptic weights. Data is divided into two sets for training, and testing. The average success rate of the method on a testing set of proteins was 90.64% in training and 89.13% in testing on three types of secondary structure a-helix, β-sheet, and coil, with correct identification coefficients of . These quality indices are all compatible with those of previous methods. From computational experiments on real and artificial structures that no method based solely on local information in the protein sequence is likely to produce significantly better results for proteins