Search

Bacterial virulence can be attributed to a wide variety of factors including toxins that harm the host. Pore-forming toxins are one class of toxins that confer virulence to the bacteria and are one of the promising targets for therapeutic intervention. In this work, we develop a sequence-based machine learning framework for the prediction of pore-forming toxins. For this, we have used distributed representation of the protein sequence encoded by reduced alphabet schemes based on conformational similarity and hydropathy index as input features to Support Vector Machines (SVMs). The choice of conformational similarity and hydropathy indices is based on the functional mechanism of pore-forming toxins. Our methodology achieves about 81% accuracy indicating that conformational similarity, an indicator of the flexibility of amino acids, along with hydrophobic index can capture the intrinsic features of pore-forming toxins that distinguish it from other types of transporter proteins. Increased understanding of the mechanisms of pore-forming toxins can further contribute to the use of such “mechanism-informed” features that may increase the prediction accuracy further.

Human Serum Albumin (HSA) has been suggested to be an alternate biomarker to the existing Hemoglobin-A1c (HbA1c) marker for glycemic monitoring. Development and usage of HSA as an alternate biomarker requires the identification of glycation sites, or equivalently, glucose-binding pockets. In this work, we combine molecular dynamics simulations of HSA and the state-of-art machine learning method Support Vector Machine (SVM) to predict glucose-binding pockets in HSA. SVM uses the three dimensional arrangement of atoms and their chemical properties to predict glucose-binding ability of a pocket. Feature selection reveals that the arrangement of atoms and their chemical properties within the first 4A from the centroid of the pocket play an important role in the binding of glucose. With a 10-fold cross validation accuracy of 84 percent, our SVM model reveals seven new potential glucose-binding sites in HSA of which two are exposed only during the dynamics of HSA. The predictions are further corroborated using docking studies. These findings can complement studies directed towards the development of HSA as an alternate biomarker for glycemic monitoring.

Accurate classification of instances depends on identification and removal of redundant features. Classification of data having high dimensionality is usually performed in conjunction with an appropriate feature selection method. Feature selection enables identification of the most informative feature subset from the enormously vast search space that can accurately classify the given data. We propose an ant colony optimization (ACO)/random forest based hybrid filter-wrapper search technique, which traverses the search space and selects a feature subset with high classifying ability. We evaluate the performance of our algorithm on four widely studied CoEPrA (Comparative Evaluation of Prediction Algorithms, http://coepra.org) datasets. The performance of the software ants mediated hybrid filter/wrapper approach compares well with the available competition results. Thus, the proposed Ant Colony Optimization based technique can effectively find small feature subsets capable of classifying with a very good accuracy and can be employed for feature subset selection with a high level of confidence.

The objective of this study was to develop hybrid genetic algorithm−support vector regression (GA-SVR)-based correlations for overall gas hold-up (ϵG), volumetric mass-transfer coefficient (kLa), and effective interfacial area (a) in bubble column reactors for gas−liquid systems employing viscous Newtonian and non-Newtonian systems as the liquid phase. The hybrid GA-SVR is a novel technique based on the feature generation approach using genetic algorithm (GA). In the present study, GA has been used for nonlinear rescaling of attributes. These, exponentially scaled, are eventually subjected to SVR training. The technique is an extension of conventional SVR technique, showing relatively enhanced results. For this purpose an extensive literature search was done. From the published literature, 1629 data points for viscous Newtonian and 845 data points for viscous non-Newtonian systems for ϵG, 500 data points for viscous Newtonian and 556 data points for viscous non-Newtonian systems for kLa, and 208 data poin...

The objective of this study was to develop a unified correlation for the volumetric mass-transfer coefficient (k L a) and effective interfacial area (a) in bubble columns for various gas-liquid systems using support vector regression (SVR-) based modeling technique. From the data published in the open literature, 1600 data points from 27 open sources spanning the years 1965-2007 for k L a and 1330 data points from 28 open sources spanning the years 1968-2007 for a were collected. Generalized SVR-based models were developed for the relationship between k L a (and a) and each design and operating parameters such as column and sparger geometry, gas-liquid physical properties, operating temperature, pressure, superficial gas velocity, and so on. Further, these models for k L a and a are available online at http://www.esnips.com/web/UICT-NCL. The proposed generalized SVR-based correlations for k L a and a have prediction accuracies of 99.08% and 98.6% and average absolute relative errors (AAREs) of 7.12% and 5.01%, respectively. Also, the SVR-based correlation provided much improved predictions compared to those obtained using empirical correlations from the literature.

Recurrence quantification analysis (RQA) has emerged as a useful tool for detecting singularities in non-stationary time-series data. In this paper, we use RQA to analyze the velocity-time data acquired using laser doppler anemometry (LDA) signals in a bubble column reactor for Single point and Multipoint point spargers. The recurring dynamical states within the velocity-time-series occurring due to the bubble and the liquid passage at the point of measurement, are quantified by RQA features (namely % Recurrence, % Determinism, % Laminarity and Entropy), which in turn are regressed using support vector regression (SVR) to predict the point gas hold-up values. It has been shown that SVR-based model for the bubble column reactor can be potentially useful for online prediction and monitoring of the point gas hold-up for different sparging conditions.

Xylanases are an industrially important class of hydrolytic enzymes that degrade xylans. Production of xylanase from a fungal culture by submerged fermentation and optimization of the operating conditions for maximum activity are the two aims of the present study. Penicillium janthinellum NCIM 1169 with Mandels-Weber medium, sugarcane bagassse (40#) as a carbon source and beef extract as a nitrogen source were used in the experiments. We did 41 experiments to see the effect of variations in carbon, nitrogen source, pH, and inoculum on xylanase activity. This data was then used to build an input/output model using multiple linear regression, back propagation neural network and lazy learning algorithm. It was found that lazy learning model correlated well in mapping input/output data. This model was then utilized as an objective function in genetic algorithm to find the optimal combination of the operating conditions to get the maximum xylanase activity. It was observed that with carbon source, 1.63%, nitrogen source, 0.16%, pH, 4.1, and inoculum, 5.5%, maximum xylanase activity of 28.98 ± 1.73 U/ml was achieved.

A hybrid methodology, viz., multicanonical jump walk annealing assisted by tabu list (MJWAT) is proposed for solving dynamic optimization problems in chemically reacting systems. This method combines the power of multicanonical sampling with the beneficial features of simulated annealing. Incorporating tabu list further enhances the efficiency of the method. The superior performance of the MJWAT is highlighted with the help of five benchmark case studies.

The objective of this study was to develop a unified data-driven correlation for the overall gas hold-up for various gas-liquid systems using support vector regression (SVR)-based modeling technique. Over the years, researchers have amply quantified the hydrodynamics of bubble column reactors in terms of the overall gas hold-up. In this work, about 1810 experimental points were collected from 40 open sources spanning the years 1965-2007. The model for overall gas hold-up was established as a function of several parameters which include superficial gas velocity, superficial liquid velocity, gas density, molecular weight of gas, sparger type, sparger hole diameter, number of sparger holes, liquid viscosity, liquid density, liquid surface tension, operating temperature, operating pressure and column diameter of the gas-liquid system. For understanding the hold-up behavior, the data used for training the model was grouped into various gas-liquid systems viz., air-water, gas-aqueous viscous liquids, gas-organic liquids, gas-aqueous electrolyte solutions and gas-liquid systems operated over a wide range of pressure. A generalized model established using SVR was evaluated for its performance for various gas-liquid systems. Statistical analysis showed that the proposed generalized SVR-based correlation for overall gas hold-up has prediction accuracy of 97% with average absolute relative error (% AARE) of 12.11%. A comparison of this correlation with the selected system specific correlations in the literature showed that the developed SVR-based correlation significantly gives enhanced prediction of overall gas hold-up.

In this work, we integrate a non-linear signal analysis method, recurrence quantification analysis (RQA), with the well-known machine-learning algorithm, support vector machines for the binary classification of protein sequences. Two different classification problems were selected, discriminating between aggregating and non-aggregating proteins and mostly disordered and completely ordered proteins, respectively. It has also been shown that classification performance of SVM models improve on selection of the most informative RQA descriptors as SVM input features.

Identification of Nuclear protein localization assumes significance as it can provide in depth insight for genome regulation and function annotation of novel proteins. A multiclass SVM classifier with various input features was employed for nuclear protein compartment identification. The input features include factor solution scores and evolutionary information (position specific scoring matrix (PSSM) score) apart from conventional dipeptide composition and pseudo amino acid composition. All the SVM classifiers with different sets of input features performed better than the previously available prediction classifiers. The jack-knife success rate thus obtained on the benchmark dataset constructed by Shen and Chou [Shen, H.B., Chou, K.C., 2005, Predicting protein subnuclear location with optimized evidence-theoretic K-nearest classifier and pseudo amino acid composition. Biochem. Biophys. Res. Commun. 337, 752-756] is 71.23%, indicating that the novel pseudo amino acid composition approach with PSSM and SVM classifier is very promising and may at least play a complimentary role to the existing methods.

In this study we have developed a support vector regression (SVR) based data-driven model for predicting two important design parameters of surface aerators, namely, the volumetric mass transfer coefficient (kLa) and fractional gas hold-up (eG). The dynamical state of the surface aerator system was captured by acquiring pressure fluctuation signals (PFSs) at various design and operating conditions. The most informative features from PFS were extracted using the chaos analysis technique, which includes estimation of Lyapunov exponent, correlation dimensions, and Kolmogorov entropy. At similar conditions the values of kLa and eG were also measured. Two different SVR models for predicting the volumetric mass transfer coefficient (kLa) and overall gas hold-up (eG) as a function of chaotic invariants, design parameters, and operating parameters were developed showing test accuracies of 98.8% and 97.1%, respectively. Such SVM based models for the surface aerator can be potentially useful on a commercial scale f...

This paper proposes PSACO (particle swarm ant colony optimization) algorithm for highly non-convex optimization problems. Both particle swarm optimization (PSO) and ant colony optimization (ACO) are co-operative, population-based global search swarm intelligence metaheuristics. PSO is inspired by social behavior of bird flocking or fish schooling, while ACO imitates foraging behavior of real life ants. In this study, we explore a simple pheromone-guided mechanism to improve the performance of PSO method for optimization of multimodal continuous functions. The proposed PSACO algorithm is tested on several benchmark functions from the usual literature. Numerical results comparisons with different metaheuristics demonstrate the effectiveness and efficiency of the proposed PSACO method.

A new method for model selection in prediction of time series is proposed. Apart from the conventional criterion of minimizing RMS error, the method also minimizes the error on the distribution of singularities, evaluated through the local Holder estimates and its probability density spectrum. Predictions of two simulated and one real time series have been done using kernel principal component regression (KPCR) and model parameters of KPCR have been selected employing the proposed as well as the conventional method. Results obtained demonstrate that the proposed method takes into account the sharp changes in a time series and improves the generalization capability of the KPCR model for better prediction of the unseen test data.

A novel method employing a unique combination of wavelet based local singularity analysis and support vector machines (SVM) classification is described and illustrated by considering the case example of flow regime identification in gas-liquid stirred tank equipped with Rushton turbine. Pressure fluctuations time series data obtained at different operating conditions were first analyzed to obtain the distribution of local Holder exponents' estimates. The relevant features from this distribution were then used as input data to the SVM classifier. Employing this method we could classify flow regimes with 98% accuracy. The results highlight the fact that the local scaling behavior of a given regime follows a distinct pattern. Further, the singularity features can be employed by intelligent machine learning based algorithms like SVM for successful online regime identification. The method can be readily applied to the other multiphase systems like bubble column, fluidized bed, etc.

Motivation: Inclusion body formation has been a major deterrent for overexpression studies since a large number of proteins form insoluble inclusion bodies when overexpressed in Escherichia coli. The formation of inclusion bodies is known to be an outcome of improper protein folding; thus the composition and arrangement of amino acids in the proteins would be a major influencing factor in deciding its aggregation propensity. There is a significant need for a prediction algorithm that would enable the rational identification of both mutants and also the ideal protein candidates for mutations that would confer higher solubility-on-overexpression instead of the presently used trial-and-error procedures. Results: Six physicochemical properties together with residue and dipeptide-compositions have been used to develop a support vector machine-based classifier to predict the overexpression status in E.coli. The prediction accuracy is ∼72% suggesting that it performs reasonably well in predicting the propensity of a protein to be soluble or to form inclusion bodies. The algorithm could also correctly predict the change in solubility for most of the point mutations reported in literature. This algorithm can be a useful tool in screening protein libraries to identify soluble variants of proteins. Avalibility: Software is available on request from the authors. Contact: balaji@iitcb.ac.in; vk.jayaraman@ncl.res.in Supplementary information: Supplementary data are available at Bioinformatics Online web site.

Motivation: Accurate prediction of genes in genomes has always been a challenging task for bioinformaticians and computational biologists. The discovery of existence of distinct scaling relations in coding and non-coding sequences has led to new perspectives in the understanding of the DNA sequences. This has motivated us to exploit the differences in the local singularity distributions for characterization and classification of coding and non-coding sequences. Results: The local singularity density distribution in the coding and non-coding sequences of four genomes was first estimated using the wavelet transform modulus maxima methodology. Support vector machines classifier was then trained with the extracted features. The trained classifier is able to provide an average test accuracy of 97.7%. The local singularity features in a DNA sequence can be exploited for successful identification of coding and non-coding sequences. Contact: Available on request from bd.kulkarni@ncl.res.in.

The paper reports on the robust pattern classification of experimental data using a combined approach of symbolization followed by support vector machine (SVM) classification. Symbolization of data removes unwanted features such as noise whereas SVM provides the classification. The SVM parameters are tuned on-line using a genetic-quasi-Newton algorithm. Benchmark examples illustrate the proposed approach.

This work provides a methodology for the control of complex non-linear systems by compensating for the non-linear part using support vector machines (SVM) and subsequently developing simple linear feedback law for control. The method tested for the benchmark Rossler and Lorenz chaotic oscillators shows excellent performance.

This article focuses on the development of a multiobjective optimization algorithm for a tubular reactor–regenerator system with a moving deactivating catalyst. The task is to find the optimal temperature profile along the tubular reactor, catalyst recycle ratio, and the regeneration capacity for maximizing the process profit flux, selectivity, and conversion. A new heuristic technique, viz, ant colony optimization method has been employed to obtain the Pareto optimal set of solutions.

This chapter presents an overview of applications of metaheuristics to solve different real-world chemical process engineering problems over the last 30 years. The first part of this chapter describes some fundamental characteristics of metaheuristics, a class of global stochastic methods and also provides the standard description of some of the most widely used metaheuristics such as simulated annealing, tabu search, genetic algorithms, and ant colony optimization (ACO). In the second part, different practical applications of these metaheuristics related to chemical process industry are covered such as heat exchanger networks (HENs), short-term scheduling of batch processes, dynamic optimization of chemical and biochemical processes, parameter estimation, and multiobjective optimization with extensive list of references.

An annular reactor packed with matrices in which the catalysts/enzyme/microorganisms are immobilized, has been simulated and the results indicate that the overall resistance and hence the conversion depend upon the Thiele modulus and another parameter, α. This parameter characterizes the ratios of the diffusion times and the ratios of length scales of the bulk liquid phase and the solid phase in the reactor shell. Analytical solutions can be obtained for linear reactions and a simplified semi-analytic method has been used for obtaining concentration profiles for nonlinear reactions.

Batch Scheduling is an important problem, relevant to a large sector of the processing industries. Methods like Simulated Annealing have been traditionally used in dealing with scheduling problems that are combinatorially complex. In this paper the performance of four different annealing strategies, Simulated Annealing (SA), Threshold Acceptance (TA) and Multicanonical Jump Walk Annealing (MJWA) with and without window factor scheduling have been compared for small and large size problems. Criteria such as robustness of the method and mean deviation from the optimal solution reveal that the MJWA with window factor scheduling is far superior to SA, TA or MJWA without window factor scheduling.

We present the results of cellular automaton and Monte Carlo simulations for the case of CO oxidation on a catalytic surface with simultaneous adsorption, reaction, and diffusion, including the Eley-Rideal step in the reaction mechanism. The results show that the carbon monoxide diffusion on the catalytic surface does not alter the qualitative nature of phase transitions. Also for any given adsorption to diffusion ratio the cellular automaton predicts slightly lower values of the phase transition point. The automaton has considerable speed advantage and corroborates the Monte Carlo simulation results.

Shifting an oscillatory or a chaotic trajectory to the unstable steady state of a nonlinear system in the presence of stochastic or deterministic load disturbances continues to be a nontrivial task. In the present work, two effective strategies for such control needs are presented. The control laws employed do not contain the process model parameters explicitly. The suggested strategies are demonstrated on two simulated nonlinear reaction systems exhibiting multi-stationarity, limit cycle oscillations, and chaos.

A simple algorithm for solving the hollow-fiber bioreactor design equations has been presented. This algorithm is quite general and is applicable to any nonlinear reaction occurring in the reactor spongy matrix. This can also be applied to other reactor types having similar configurations, like the wall-coated enzyme reactors.

Pseudo-periodic signals are rampant in biomedical applications but are difficult to analyze. One approach is to compute time domain parameters of each individual cycle in the pseudo-periodic signal. The classic approach requires repeated computation in each cycle, which tends to be either error prone, computationally burdensome, or requires manual effort. We provide a novel combination of the pitch synchronous wavelet transform which when combined with dynamic time warping results in effective quantification of cycles in the pseudo-periodic signal. We demonstrate our application of this method in studying the arterial pulse. The results show that our approach is feasible and effective, and confirms further scope in other applications.

Cellular automata (CA) rules for a model surface reaction with Eley-Rideal step (exemplifying CO oxidation) have been formulated. The CA simulations corroborate the earlier known Monte Carlo (MC) simulation results and show qualitative and near quantitative resemblance. The results are analyzed and discussed in the light of MC and mean field solutions.

The mass balance equations for hollow fibre bioreactors have been solved for the zero-order limit of the Michaelis-Menten kinetics. As in the case of first-order reactions, the membrane and spongy matrix equations can be decoupled from the overall set of equations. For the case of the substrate remaining constant everywhere in the reactor a solution in terms of hypergeometric series is possible. For the substrate exhaustion case the extinction radius is a non-linear function of the system parameters and thus the equations have to be solved by numerical methods.

Automatically classifying ECG recordings for Malignant Ventricular Arrhythmia is fraught with several difficulties. Even normal ECG signals exhibit only quasi-periodic nature, and contain various irregularities. The key to more accurate detection is the use of position, and amount of local singularities in the signals.In this paper, we propose a Holder-SVM detection algorithm using a novel hybrid arrangement of binary and multiclass SVMs designed to take care of class imbalance rampant in biomedical signals. As a result, we significantly reduce the number of false negatives – patients falsely classified as normal. We used the MIT-BIH Arrhythmia database for even different arrhythmias. We compare our hybrid SVM with a suitable conventional SVM, and show better results.We also use the new arrangement for features proposed earlier, and demonstrate the gain in accuracy. Our concept of hybrid SVM is applicable to a wide variety of multiclass classification problems.

Building accurate quantitative structure-activity relationships (QSAR) is important in drug design, environmental modeling, toxicology, and chemical property prediction. QSAR methods can be utilized to solve mainly two types of problems viz., pattern recognition, (or classification) where output is discrete (i.e. class information), e.g., active or non-active molecule, binding or non-binding molecule etc., and function approximation, (i.e. regression) where the output is continuous (e.g., actual activity prediction). The present review deals with the second type of problem (regression) with specific attention to one of the most effective machine learning procedures, viz. lazy learning. The methodologies of the algorithm along with the relevant technical information are discussed in detail. We also present three real life case studies to briefly outline the typical characteristics of the modeling formalism.

Extensive research has been done to show that heartbeats are composed of the interaction of many physiological components operating on different time scales, with nonlinear and self-regulating nature. The more direct, and easily accessible manifestation of the heartbeat is the pulse; however, it has not been studied anywhere near as extensively. In this paper, we establish the relevance of the multi-fractal formalism for the arterial pulse, which has long been used as a fundamental tool for diagnosis in the Traditional Indian Medicine, (Ayurveda). The finding of power-law correlations through detrended fluctuation analysis indicates presence of scale-invariant, fractal structures in the pulse. These fractal structures are then further established by self-affine cascades of beat-to-beat fluctuations revealed by wavelet decomposition at different time scales. Finally, we investigate how these pulse dynamics change with age, and disorder. The analytic tools we discuss may be used on a wide range of physiological signals.

Heart rate variability (HRV) provides an estimate of sympathetic and parasympathetic influences on the heart rate. Although HRV has been extensively studied, sustained clinical use is still outstanding. The noninvasive, convenient, and inexpensive arterial pulse originate from heartbeats, but has not been studied in a systematic fashion except in rudimentary ways. In this paper, we present pulse rate variability (PRV) as an alternative to HRV. We give evidence for the detection of disorders in patients using PRV, paving the way for future clinical use.

It is well known that Dynamic Time Warping (DTW) is superior to Euclidean distance as a similarity measure in time series analyses. Use of DTW with the recently introduced warping window constraints and lower bounding measures has significantly increased the accuracy of time series classification while reducing the computational expense required. The warping window technique learns arbitrary constraints on the warping path while performing time series alignment. This work utilizes genetic algorithms to find the optimal warping window constraints which provide a better classification accuracy. Performance of the proposed methodology has been investigated on two problems from diverse domains with favorable results.

Owing to the superiority of Dynamic Time Warping as a similarity measure of time series, it can become an effective tool for fault diagnosis in chemical process plants. However, direct application of Dynamic Time Warping can be computationally inefficient, given the complexity involved. In this work we have tackled this problem by employing a warping window constraint and a Lower Bounding measure. A novel methodology for online fault diagnosis with Dynamic Time Warping has been suggested and its performance has been investigated using two simulated case studies.

We employed a granular support vector Machines(GSVM) for prediction of soluble proteins on over expression in Escherichia coli. Granular computing splits the feature space into a set of subspaces (or information granules) such as classes, subsets, clusters and intervals [14]. By the principle of divide and conquer it decomposes a bigger complex problem into smaller and computationally simpler problems. Each of the granules is then solved independently and all the results are aggregated to form the final solution. For the purpose of granulation association rules was employed. The results indicate that a difficult imbalanced classification problem can be successfully solved by employing GSVM.