Whole Genome Sequencing (WGS) provides information for each base of the entire 3.2 billion base pairs of the diploid human genome. Therefore, WGS plays an important role in identifying genetic ...variations for populations and understanding disease signatures in cohort studies or cases with rare genetic disorders. Nonetheless, discoveries from high throughput WGS are dependent on efficient processing, analyzing, and storing this enormous amount of genomic sequencing data, often in the scale of petabytes. Although there has been a significant reduction in genome sequencing costs in recent years, high-performance computation costs have not decreased in a directly proportional fashion.
The objective of the present work is to develop a Docker-based container method for human whole genome sequencing data processing and analysis for detecting genetic variations from paired end WGS short reads. Our method provides an approach to simultaneously process multiple genomes within a single compute system while guaranteeing sustained and stable handling of the memory requirements for the genomic data processing and ensuring no unwanted termination of the currently running parallel jobs. This method also achieves a 40 % reduction in execution time. To encourage widespread adoption and ease of WGS analysis, our containerized pipeline will be made publicly available. We have tested this approach for human genome data from Illumina WGS platforms and report the benchmark metrics in two different workstation environments in this communication. Compared to truth sets, our approach calls variants with 99 % precision and recall.
The present paper introduces the particle swarm optimization (PSO) technique to develop an efficient forecasting model for prediction of various stock indices. The connecting weights of the adaptive ...linear combiner based model are optimized by the PSO so that its mean square error(MSE) is minimized. The short and long term prediction performance of the model is evaluated with test data and the results obtained are compared with those obtained from the multilayer perceptron (MLP) based model. It is in general observed that the proposed model is computationally more efficient, prediction wise more accurate and takes less training time compared to the standard MLP based model.
The objective of portfolio optimization is to find an optimal set of assets to invest on, as well as to determine the optimal investment for each asset. This optimal selection and weighting of assets ...is a multi-objective problem where total profit of investment has to be maximized and total risk is to be minimized. In this paper the portfolio optimization problem is solved using three different multi-objective algorithms and their performance have been compared in terms of Pareto fronts, the delta, C and S metrics. Exhaustive simulation study of various portfolios clearly demonstrates the superior portfolio management capability of non-dominated sorting genetic algorithm II (NSGA II) based method compared to other two methods.
Adversarial input detection has emerged as a prominent technique to harden Deep Neural Networks (DNNs) against adversarial attacks. Most prior works use neural network-based detectors or complex ...statistical analysis for adversarial detection. These approaches are computationally intensive and vulnerable to adversarial attacks. To this end, we propose DetectX-A hardware friendly adversarial detection mechanism using hardware signatures like Sum of column Currents (SoI) in memristive crossbars (XBar). We show that adversarial inputs have higher SoI compared to clean inputs. However, the difference is too small for reliable adversarial detection. Hence, we propose a dual-phase training methodology: Phase1 training is geared towards increasing the separation between clean and adversarial SoIs; Phase2 training improves the overall robustness against different strengths of adversarial attacks. For hardware-based adversarial detection, we implement the DetectX module using 32 nm CMOS circuits and integrate it with a Neurosim-like analog crossbar architecture. We perform hardware evaluation of the Neurosim+DetectX system on the Neurosim platform using datasets-CIFAR10(VGG8), CIFAR100(VGG16) and TinyImagenet(ResNet18). Our experiments show that DetectX is 10x-25x more energy efficient and immune to dynamic adversarial attacks compared to previous state-of-the-art works. Moreover, we achieve high detection performance (ROC-AUC>0.95) for strong white-box and black-box attacks.
Social networks have emerged as efficient platforms to connect people worldwide and facilitate the rapid spread of information. Identifying influential nodes in social networks to accelerate the ...spread of particular information across the network formulates the study of influence maximization (IM). In this paper, inspired by deep learning techniques, we propose a novel approach to solve the influence maximization problem as a classical regression task using transfer learning via graph-based long short-term memory (GLSTM). We start by calculating three popular node centrality methods as feature vectors for the nodes in the network and every node’s individual influence under susceptible–infected–recovered (SIR) information diffusion model, which forms the labels of the nodes in the network. The generated feature vectors and their corresponding labels for all the nodes are then fed into a graph-based long short-term memory (GLSTM). The proposed architecture is trained on a vast and complex network to generalize the model parameters better. The trained model is then used to predict the probable influence of every node in the target network. The proposed model is compared with some of the well-known and recently proposed algorithms of influence maximization on several real-life networks using the popular SIR model of information diffusion. The intensive experiments suggest that the proposed model outperforms these well-known and recently proposed influence maximization algorithms.
•A novel method for IM is proposed based on transfer learning via a graph-based LSTM.•Obtained features on the training network fed to GLSTM to learn the model parameters.•The trained GLSTM model predicts the spreading influence of nodes of target network.•Simulations on various datasets reveal the improved performance of the proposed work.
With the boom in technologies and mobile networks in recent years, online social networks have become an integral part of our daily lives. These virtual networks connect people worldwide and provide ...them excellent platforms for promoting their products and ideas. It is often the case that certain users are more influential than others present on social networks. The process of efficiently recognizing influential users to maximize a particular piece of information across a network is known as Influence Maximization (IM). In this paper, we propose a novel method of influence maximization by using the idea of graph embedding and graph neural networks. This study intends to convert the problem of influence maximization in complex networks into a pseudo-regression problem. As part of our approach, first, we employ the struc2vec node embedding to generate the embedding for every node in the network, and the obtained embedding acts as the feature for each node. The nodes and their features are then fed into a Graph Neural Network (GNN) based regressor. The labels required for training the GNN for the regression task are obtained by calculating every node’s influence under the Susceptible-Infected-Recovered (SIR) and Independent Cascade (IC) information diffusion model. After that, we select an optimal training network by performing a parametric analysis on synthetic test networks. Finally, the trained model is used to predict the probable influence of nodes on the target network. Experimental results on several real-life networks based on various evaluation metrics show that the proposed approach outperforms some of the classical and recently proposed influence maximization methods.
In today's competitive business environment, ensuring the sustainability in long-run remains a key challenge for business organizations. Using fossil fuel-based technologies for in carrying out ...operating activities may put organizational sustainability at stake. Until now, researcher have not paid much attention to explore these issues. Consequently, following the recommendations of natural-resource-based view current study attempts to identify the issues & sub-issues concerning fossil-fuel usage by businesses and solutions to mitigate the effect of these. To understand the priority order of these identified issues and respective sub-issues, BWM (Best Worst Method) is employed. Further, current work uses VIKOR (VlseKriterijumska Optimizacija I Kompromisno Resenje) to explore the ranks of the proposed solutions considering the issues under consideration. Current analysis indicates “Legislation” to remain the most important issue followed by “Environment.” Further, ‘Sensitising and training employees towards usage of fossil fuel-based technologies with no wastage or minimum wastage’ is observed as the most workable solution in the short-term. The solution “Switching to renewable energy sources” is ranked on the second position. Findings of current study carry significant informational content that may assist organizations in developing the strategic framework to achieving long-term sustainability.
Transformers have revolutionized various real-world applications from natural language processing to computer vision. However, traditional von-Neumann computing paradigm faces memory and bandwidth ...limitations in accelerating transformers owing to their massive model sizes. To this end, In-memory Computing (IMC) crossbars based on Non-volatile Memories (NVMs), due to their ability to perform highly parallelized Matrix-Vector-Multiplications (MVMs) with high energy-efficiencies, have emerged as a promising solution for accelerating transformers. However, analog MVM operations in crossbars introduce non-idealities, such as stochastic read & write noise, which affect the inference accuracy of the deployed transformers. Specifically, we find pre-trained Vision Transformers (ViTs) to be vulnerable on crossbars due to the impact of write noise on the dynamically-generated Key (K) and Value (V) matrices in the attention layers, an effect not accounted for in prior studies. We, thus, propose ClipFormer, a transformation on the K and V matrices during inference, to boost the non-ideal accuracies of pre-trained ViT models. ClipFormer requires no additional hardware and training overhead and is amenable to transformers deployed on any memristive crossbar platform impacted by write non-idealities. Our experiments on Imagenet-1k dataset using pre-trained DeiT-S transformers, subjected to standard training and variation-aware-training, show >10-40% higher non-ideal accuracies at the high write noise regime by applying ClipFormer.
•An evolving Fuzzy-PID controller for the trajectory control of a 7-DOF manipulator.•The scaling factors of the fuzzy controller are evolved by using meta-heuristics.•The proposed technique is more ...efficient in terms of reduced tracking error.•The proposed control design is also more robust to external noise and disturbances.•This method relies lesser on the system dynamics and is easily implementable.
Articulated manipulators constitute a major fraction of industrial robots. With the increase in the need to perform complex tasks requiring a higher degree of accuracy, such manipulators require a more robust and efficient control. In this paper, an evolving Fuzzy-PID control design for the trajectory tracking problem of a redundant 7-DOF serial manipulator is proposed. The scaling factors of the Fuzzy controller are evolved over time by using recent meta-heuristic techniques, namely, Moth-Flame Optimisation (MFO), Novel Bat Algorithm (NBA) and Sine Cosine Algorithm (SCA). The proposed technique which utilizes Intelligent control methods combined with meta-heuristics provides a promising hybrid control design, which has been found to be more efficient in our software implementation, in terms of reduced tracking error of the manipulator for its kinematic control in the joint space as compared to other controllers of the same nature that are generally used in practise. The performance of the controller is compared in the time-domain responses under external disturbances and have been found to outperform standard controllers without the need of any additional training.
PDF
