Predicting membrane protein types by incorporating a novel feature set into Chou's general PseAAC.

Membrane proteins are vital type of proteins that serve as channels, receptors and energy transducers in a cell. They perform various important functions, which are mainly associated with their types. They are also attractive targets of drug discovery for various diseases. So predicting membrane protein types is a crucial and challenging research area in bioinformatics and proteomics. Because of vast investigation of uncharacterized protein sequences in databases, customary biophysical techniques are extremely tedious, costly and vulnerable to mistakes. Subsequently, it is very attractive to build a vigorous, solid, proficient technique to predict membrane protein types. In this work, a novel feature set Exchange Group Based Protein Sequence Representation (EGBPSR) is proposed for classification of membrane proteins with two new feature extraction strategies known as Exchange Group Local Pattern (EGLP) and Amino acid Interval Pattern (AIP). Imbalanced dataset and large dataset are often handled well by decision tree classifiers. Since imbalanced dataset are taken, the performance of various decision tree classifiers such as Decision Tree (DT), Classification and Regression Tree (CART), ensemble methods such as Adaboost, Random Under Sampling (RUS) boost, Rotation forest and Random forest are analyzed. The overall accuracy achieved in predicting membrane protein types is 96.45%.

Predicting membrane protein types using various decision tree classifiers based on various modes of general PseAAC for imbalanced datasets.

Predicting membrane protein types is an important and challenging research area in bioinformatics and proteomics. Traditional biophysical methods are used to classify membrane protein types. Due to large exploration of uncharacterized protein sequences in databases, traditional methods are very time consuming, expensive and susceptible to errors. Hence, it is highly desirable to develop a robust, reliable, and efficient method to predict membrane protein types. Imbalanced datasets and large datasets are often handled well by decision tree classifiers. Since imbalanced datasets are taken, the performance of various decision tree classifiers such as Decision Tree (DT), Classification And Regression Tree (CART), C4.5, Random tree, REP (Reduced Error Pruning) tree, ensemble methods such as Adaboost, RUS (Random Under Sampling) boost, Rotation forest and Random forest are analysed. Among the various decision tree classifiers Random forest performs well in less time with good accuracy of 96.35%. Another inference is RUS boost decision tree classifier is able to classify one or two samples in the class with very less samples while the other classifiers such as DT, Adaboost, Rotation forest and Random forest are not sensitive for the classes with fewer samples. Also the performance of decision tree classifiers is compared with SVM (Support Vector Machine) and Naive Bayes classifier.

Firefly algorithm based WSN-IoT security enhancement with machine learning for intrusion detection.

A Wireless Sensor Network (WSN) aided by the Internet of Things (IoT) is a collaborative system of WSN systems and IoT networks are work to exchange, gather, and handle data. The primary objective of this collaboration is to enhance data analysis and automation to facilitate improved decision-making. Securing IoT with the assistance of WSN necessitates the implementation of protective measures to confirm the safety and reliability of the interconnected WSN and IoT components. This research significantly advances the current state of the art in IoT and WSN security by synergistically harnessing the potential of machine learning and the Firefly Algorithm. The contributions of this work are twofold: firstly, the proposed FA-ML technique exhibits an exceptional capability to enhance intrusion detection accuracy within the WSN-IoT landscape. Secondly, the amalgamation of the Firefly Algorithm and machine learning introduces a novel dimension to the domain of security-oriented optimization techniques. The implications of this research resonate across various sectors, ranging from critical infrastructure protection to industrial automation and beyond, where safeguarding the integrity of interconnected systems are of paramount importance. The amalgamation of cutting-edge machine learning and bio-inspired algorithms marks a pivotal step forward in crafting robust and intelligent security measures for the evolving landscape of IoT-driven technologies. For intrusion detection in the WSN-IoT, the FA-ML method employs a support vector machine (SVM) machine model for classification with parameter tuning accomplished using a Grey Wolf Optimizer (GWO) algorithm. The experimental evaluation is simulated using NSL-KDD Dataset, revealing the remarkable enhancement of the FA-ML technique, achieving a maximum accuracy of 99.34%. In comparison, the KNN-PSO and XGBoost models achieved lower accuracies of 96.42% and 95.36%, respectively. The findings validate the potential of the FA-ML technique as an active security solution for WSN-IoT systems, harnessing the power of machine learning and the Firefly Algorithm to bolster intrusion detection capabilities.

Segmentation and classification of bright lesions to diagnose diabetic retinopathy in retinal images.

In view of predicting bright lesions such as hard exudates, cotton wool spots, and drusen in retinal images, three different segmentation techniques have been proposed and their effectiveness is compared with existing segmentation techniques. The benchmark images with annotations present in the structured analysis of the retina (STARE) database is considered for testing the proposed techniques. The proposed segmentation techniques such as region growing (RG), region growing with background correction (RGWBC), and adaptive region growing with background correction (ARGWBC) have been used, and the effectiveness of the algorithms is compared with existing fuzzy-based techniques. Images of eight categories of various annotations and 10 images in each category have been used to test the consistency of the proposed algorithms. Among the proposed techniques, ARGWBC has been identified to be the best method for segmenting the bright lesions based on its sensitivity, specificity, and accuracy. Fifteen different features are extracted from retinal images for the purpose of identification and classification of bright lesions. Feedforward backpropagation neural network (FFBPNN) and pattern recognition neural network (PRNN) are used for the classification of normal/abnormal images. Probabilistic neural network (PNN), radial basis exact fit (RBE), radial basis fewer neurons (RB), and FFBPNN are used for further bright lesion classification and achieve 100% accuracy.