Article
Biochemical Research Methods
Muhammad Awais, Waqar Hussain, Nouman Rasool, Yaser Daanial Khan
Summary: A novel and efficient model for predicting tumor suppressor proteins was proposed in this study. The model showed excellent performance in accuracy, sensitivity, and specificity, with potential for further improvements in computational methods.
CURRENT BIOINFORMATICS
(2021)
Article
Engineering, Biomedical
Khalid Allehaibi, Yaser Daanial Khan, Sher Afzal Khan
Summary: The process of angiogenesis plays a vital role in controlling the growth of blood vessels within tissues, with angiogenesis proteins being crucial for this process. The balance of signals is essential for proper angiogenesis functioning, as imbalance can lead to abnormal growth or diseases like cancer. The study focuses on developing a prediction model using various classifiers to identify angiogenesis proteins and predict their association with tumor angiogenesis, with the model's performance evaluated through different validation techniques.
APPLIED BIONICS AND BIOMECHANICS
(2021)
Article
Multidisciplinary Sciences
Yaser Daanial Khan, Nabeel Sabir Khan, Sheraz Naseer, Ahmad Hassan Butt
Summary: Sumoylation is a post-translational modification crucial for cell adaption and protein function, with implications in various human diseases. Predicting Sumoylation sites is significant, and a new technique based on statistical features shows promising accuracy and performance.
Article
Biochemical Research Methods
Muhammad Awais, Waqar Hussain, Yaser Daanial Khan, Nouman Rasool, Sher Afzal Khan, Kuo-Chen Chou
Summary: Protein phosphorylation is a key mechanism in cells, but predicting phosphohistidine sites accurately is challenging. The proposed iPhosH-PseAAC model successfully achieves high accuracy in predicting phosphohistidine sites, providing a new approach for studying biological processes.
IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS
(2021)
Article
Multidisciplinary Sciences
Sheraz Naseer, Rao Faizan Ali, Amgad Muneer, Suliman Mohamed Fati
Summary: Amidation is an important post-translational modification in peptides, with amidated peptides being less susceptible to degradation and having extended half-lives. The use of deep neural networks and Pseudo Amino Acid Compositions can efficiently and accurately identify valine amidation sites, with the Convolutional neural network-based model showing the best performance. This proposed model can supplement in-vivo methods and enhance understanding of valine amidation in various biological processes.
Article
Multidisciplinary Sciences
Hima Elsa Shaji, Arun K. Tangirala, Lelitha Vanajakshi
Summary: This study proposes an iterative joint clustering and prediction approach to improve the accuracy of travel time predictions. By creating data clusters that are sensitive to the quality of predictions, this method has been validated in real-world traffic scenarios.
Article
Biology
Alaa Omran Almagrabi, Yaser Daanial Khan, Sher Afzal Khan
Summary: Phosphoaspartate plays a crucial role in signal transduction, and iPhosD-PseAAC is a new computational model that uses multiple techniques and artificial neural networks for predicting phosphoaspartate sites with high accuracy and efficiency.
Article
Polymer Science
Francisco M. Monticeli, Roberta M. Neves, Heitor L. Ornaghi Jr, Jose Humberto S. Almeida
Summary: The study focuses on the fabrication of high-performance 3D printable CF/epoxy composites, using approaches based on artificial neural networks, analysis of variance, and response surface methodology for data prediction and analysis. The predicted results show high reliability and low error level, approaching experimental results. Various parameters influencing the fabrication of the composites are considered, and fast and streamlined fabrications of different composite materials with tailor-made properties are demonstrated.
Article
Chemistry, Analytical
Galina Malykhina, Dmitry Tarkhov, Viacheslav Shkodyrev, Tatiana Lazovskaya
Summary: Efficient measurement of LED efficiency is crucial for mass production automation, and current methods involve comparing LED crystal cooling curves and utilizing blind signal extraction based on neural networks. Statistical analysis reveals that the signal and noise have different forms of probability density functions, so using generalized moments for signal extraction is appropriate.
Article
Physics, Fluids & Plasmas
Frederieke Richert, Roman Worschech, Bernd Rosenow
Summary: In the context of overparametrized deep neural networks, the student network may have a larger expressivity than the data generating process. In the student-teacher scenario, if the student network has more hidden units than the teacher, evidence suggests that the approach to perfect learning occurs in a power-law fashion rather than exponentially, with all student nodes able to replicate one of the teacher nodes if suitable rescaling is applied to the outputs and the numbers of hidden units are commensurate.
Article
Computer Science, Artificial Intelligence
Jitendra Kumar, Ashutosh Kumar Singh
Summary: This research aims to investigate the performance of nature-inspired metaheuristic algorithms on workload prediction in a cloud environment, with the Blackhole Algorithm (BhA) showing promising results in predictive accuracy.
APPLIED SOFT COMPUTING
(2021)
Article
Computer Science, Interdisciplinary Applications
Meetesh Nevendra, Pradeep Singh
Summary: Software defect prediction (SDP) is the process of developing a model to distinguish defective modules or classes in the early stages of software development. Recent research has focused on using deep learning (DL) techniques for SDP and has shown that DL techniques achieve better results than traditional machine learning approaches in terms of both accuracy and significance. However, there is still a lack of sufficient utilization and further investigation is needed to obtain well-formed and generalizable results.
ARCHIVES OF COMPUTATIONAL METHODS IN ENGINEERING
(2022)
Review
Computer Science, Information Systems
Zengchen Yu, Ke Wang, Zhibo Wan, Shuxuan Xie, Zhihan Lv
Summary: Due to its automatic feature learning ability and high performance, deep learning has gradually become the mainstream of artificial intelligence in recent years, playing a role in many fields. This paper introduces several deep learning algorithms such as Artificial Neural Network (NN), FM-Deep Learning, Convolutional NN and Recurrent NN, and explains their theory, development history, and applications in disease prediction. The paper also analyzes the current defects in the disease prediction field and provides some current solutions. Furthermore, it discusses two major trends in the future disease prediction and medical field - integrating Digital Twins and promoting precision medicine.
CLUSTER COMPUTING-THE JOURNAL OF NETWORKS SOFTWARE TOOLS AND APPLICATIONS
(2023)
Article
Ecology
Zhiqiang Zheng, Hao Ding, Zhi Weng, Lixin Wang
Summary: This study proposes a new water quality prediction model that achieves high-precision multiparameter analysis on small data samples. It is of great significance for the protection and management of water environments.
ECOLOGICAL INFORMATICS
(2023)
Article
Mechanics
Murat Kara, Abdullah Secgin, Tuba Baygun, Cagri Gokhan Akyol
Summary: This study verifies the applicability of Statistical Moment (SM) based modelling method in uncertainty modelling of realistic structures by employing high-degree statistical moments in stochastic equations. Experimental and numerical results show that this approach can effectively be used in uncertainty modelling of realistic structures.
COMPOSITE STRUCTURES
(2022)
Article
Biochemical Research Methods
Waqar Hussain
Summary: This paper proposes a method for predicting short antimicrobial peptides using deep neural networks and compares it with previous methods. The results show that this method can accurately and efficiently identify antimicrobial peptides.
BRIEFINGS IN BIOINFORMATICS
(2022)
Article
Automation & Control Systems
Muhammad Arif, Saeed Ahmed, Fang Ge, Muhammad Kabir, Yaser Daanial Khan, Dong-Jun Yu, Maha Thafar
Summary: A novel predictor called Stack-ACPred has been developed for the accurate identification of anticancer peptides (ACPs). This method combines three feature encoding strategies and utilizes an optimization algorithm for feature fusion and attribute selection, resulting in the construction of an effective ensemble model. Empirical results demonstrate the excellent discriminative power of this method for annotating large scale ACPs.
CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS
(2022)
Article
Biochemical Research Methods
Sharaf J. Malebary, Ebraheem Alzahrani, Yaser Daanial Khan
Summary: This study aims to accurately predict glutamine sites vulnerable to methylation using computationally intelligent classifiers, with deep learning performing best among them.
JOURNAL OF MOLECULAR GRAPHICS & MODELLING
(2022)
Article
Computer Science, Information Systems
Usama Hasan, Waqar Hussain, Nouman Rasool
Summary: This study presents an automated method for human identification based on ear pinna images, using deep learning and feature fusion to achieve high accuracy.
MULTIMEDIA TOOLS AND APPLICATIONS
(2022)
Article
Biochemical Research Methods
Wajdi Alghamdi, Muhammad Attique, Ebraheem Alzahrani, Malik Zaka Ullah, Yaser Daanial Khan
Summary: The identification of B-cell epitopes is crucial for therapeutics, vaccine development, and antibody production. Experimental approaches are challenging and time-consuming, leading to the development of computational methods. LBCEPred, a python-based web-tool, outperforms existing sequence-based models in predicting B-cell epitopes.
BRIEFINGS IN BIOINFORMATICS
(2022)
Article
Engineering, Biomedical
Omar Barukab, Yaser Daanial Khan, Sher Afzal Khan, Kuo-Chen Chou
Summary: In the domain of genome annotation, identifying DNA-binding proteins is a crucial challenge. Existing methods are expensive and time-consuming. This study proposes a methodology called DNAPred_Prot, which efficiently predicts DNA-binding proteins using various features from protein sequences. The results demonstrate that this methodology outperforms other methods in terms of accuracy.
APPLIED BIONICS AND BIOMECHANICS
(2022)
Article
Mathematics, Applied
Yaser Daanial Khan, M. Khalid Mahmood, Daud Ahmad, Nasser M. Al-Zidi
Summary: This article introduces an efficient and novel probabilistic technique for content-based image retrieval, utilizing glyph structure patterns to form content representations. By applying a mixture model for gamma distribution and refining parameters through maximum likelihood, the proposed method retrieves matching images with comparable distribution patterns.
JOURNAL OF FUNCTION SPACES
(2022)
Article
Green & Sustainable Science & Technology
Zengjian Huang, Amna Shahzadi, Yaser Daanial Khan
Summary: This study investigates the impact of social and technical Q4.0 on I4.0 technologies and circular economy practices in SME manufacturing enterprises. The findings suggest that Q4.0 practices significantly improve I4.0 technologies and CEP, with technical Q4.0 acting as a mediator between social Q4.0 practices, I4.0 technologies, and CEP.
Article
Medicine, General & Internal
Arfa Hassan, Tamim Alkhalifah, Fahad Alturise, Yaser Daanial Khan
Summary: This article introduces a method for predicting clear cell renal carcinoma using genomic sequences. A neural network model was trained using data from IntOgen and NCBI, and achieved high accuracy results through cross-validation and validation dataset testing.
Article
Genetics & Heredity
Asghar Ali Shah, Fahad Alturise, Tamim Alkhalifah, Amna Faisal, Yaser Daanial Khan
Summary: Cholangiocarcinoma is the leading cause of mortality and disability globally, with higher fatality rates among Asian populations. Detecting cholangiocarcinoma at an earlier stage can greatly improve treatment outcomes. An ensemble deep learning model is developed for early identification, achieving high accuracy of 98% in independent testing.
Article
Medicine, General & Internal
Ahmad Hassan Butt, Tamim Alkhalifah, Fahad Alturise, Yaser Daanial Khan
Summary: Hormone-binding proteins (HBPs) are specific carrier proteins that bind to a given hormone. A soluble carrier hormone binding protein (HBP) interacts with growth hormone, modulating or inhibiting hormone signaling. Accurate identification of HBPs is crucial for understanding their biological mechanisms, and a computational method based on machine learning is suggested. This method achieved high accuracy and demonstrated the importance of Hahn moment-based features.
Article
Engineering, Biomedical
Abdul Rafay, Waqar Hussain
Summary: Skin diseases are common and often underestimated health issues that affect nearly one-third of the global population. Deep learning algorithms have shown great potential in accurately diagnosing these diseases. This research presents a novel dataset of 31 skin diseases and uses three different CNN types for transfer learning, with EfficientNet achieving the highest testing accuracy and further fine-tuning.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2023)
Article
Medicine, General & Internal
Fahad M. Alotaibi, Yaser Daanial Khan
Summary: Mutations in genes can be detected to diagnose carcinomas, and deep learning approaches such as LSTM, Bi-LSTM, and GRU are proposed to optimize the identification of gastric carcinoma progression. The study includes 61 carcinogenic driver genes and validates the models using SCT, FCVT, and IST. IST predictions show high accuracy, sensitivity, and specificity for all three models.
Article
Ophthalmology
Abdul Rafay, Zaeem Asghar, Hamza Manzoor, Waqar Hussain
Summary: The eyes are essential for daily life, but eye diseases are often underestimated until it is too late. A novel method called EyeCNN is proposed to identify eye diseases through retinal images, offering the potential to diagnose conditions accurately and efficiently.
INTERNATIONAL OPHTHALMOLOGY
(2023)