ORI-Deep: improving the accuracy for predicting origin of replication sites by using a blend of features and long short-term memory network

Shahid, Mahwish; Ilyas, M.; Hussain, Waqar; Khan, Yaser Daanial

doi:10.1093/bib/bbac001

Cited by 16 publications

(8 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus, by equation ( 23 ), the overall accuracy, specificity, sensitivity, and MCC can be easily understood compared to the equation defined in ( 22 ) which is authenticated only forsingle-label systems. A real unique metric set is required for systems that are multilabelled as described in [ 78 ] and whose emergence is becoming common in biomedicine [ 79 ], system medicine [ 80 ], and system biology [ 81 ].…”

Section: Experiments and Resultsmentioning

confidence: 99%

“…Developing a convenient web server is the 5 th step in the five-step rule. As specified and explained in the number of recent publications [ 73 – 75 , 81 , 82 ], for development of practical, more useful forecasting methods and tools for computation in the future need a web server that is publicly available at the link and easy to use. The user can follow a series of steps to take benefit from the study using a web server.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

DNAPred_Prot: Identification of DNA-Binding Proteins Using Composition- and Position-Based Features

Barukab

Khan

et al. 2022

Applied Bionics and Biomechanics

Self Cite

View full text Add to dashboard Cite

In the domain of genome annotation, the identification of DNA-binding protein is one of the crucial challenges. DNA is considered a blueprint for the cell. It contained all necessary information for building and maintaining the trait of an organism. It is DNA, which makes a living thing, a living thing. Protein interaction with DNA performs an essential role in regulating DNA functions such as DNA repair, transcription, and regulation. Identification of these proteins is a crucial task for understanding the regulation of genes. Several methods have been developed to identify the binding sites of DNA and protein depending upon the structures and sequences, but they were costly and time-consuming. Therefore, we propose a methodology named “DNAPred_Prot”, which uses various position and frequency-dependent features from protein sequences for efficient and effective prediction of DNA-binding proteins. Using testing techniques like 10-fold cross-validation and jackknife testing an accuracy of 94.95% and 95.11% was yielded, respectively. The results of SVM and ANN were also compared with those of a random forest classifier. The robustness of the proposed model was evaluated by using the independent dataset PDB186, and an accuracy of 91.47% was achieved by it. From these results, it can be predicted that the suggested methodology performs better than other extant methods for the identification of DNA-binding proteins.

show abstract

Section: Experiments and Resultsmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

DNAPred_Prot: Identification of DNA-Binding Proteins Using Composition- and Position-Based Features

Barukab

Khan

et al. 2022

Applied Bionics and Biomechanics

Self Cite

View full text Add to dashboard Cite

show abstract

“…All the above statistical functions, used for feature extraction, have specific biological significance. 33 – 44 These functions extract information related to the position and composition of DNA gene sequences. Feature extraction helps in extracting very useful information such as the frequency of each element in DNA gene sequences, position relative to the occurrence, composition of a specific gene, and absolute position of each element of gene sequences.…”

Section: Methodsmentioning

confidence: 99%

Evaluation of deep learning techniques for identification of sarcoma-causing carcinogenic mutations

et al. 2022

Self Cite

View full text Add to dashboard Cite

The abnormal growth of human healthy cells is called cancer. One of the major types of cancer is sarcoma, mostly found in human bones and soft tissue cells. It commonly occurs in children. According to a survey of the United States of America, there are more than 17,000 sarcoma patients registered each year which is 15% of all cancer cases. Recognition of cancer at its early stage saves many lives. The proposed study developed a framework for the early detection of human sarcoma cancer using deep learning Recurrent Neural Network (RNN) algorithms. The DNA of a human cell is made up of 25,000 to 30,000 genes. Each gene is represented by sequences of nucleotides. The nucleotides in a sequence of a driver gene can change which is termed as mutations. Some mutations can cause cancer. There are seven types of a gene whose mutation causes sarcoma cancer. The study uses the dataset which has been taken from more than 134 samples and includes 141 mutations in 8 driver genes. On these gene sequences RNN algorithms Long and Short-Term Memory (LSTM), Gated Recurrent Units and Bi-directional LSTM (Bi-LSTM) are used for training. Rigorous testing techniques such as Self-consistency testing, independent set testing, 10-fold cross-validation test are applied for the validation of results. These validation techniques yield several metrics such as Area Under the Curve (AUC), sensitivity, specificity, Mathew's correlation coefficient, loss, and accuracy. The proposed algorithm exhibits an accuracy of 99.6% with an AUC value of 1.00.

show abstract

“…In machine learning (ML), the model performance is naturally assessed via cross-validation (CV). There are three tests in the research community to determine the discriminatory power of the designed framework: K-fold also called subsampling, Jackknife, i.e., leave-one-out and independent tests [ 56 , 57 ]. The Jackknife test provides exceptional and encouraging results to train a model [ 58 ]; however, the main cons are computational cast due to a large number of calculations [ 59 ].…”

Section: Proposed Framework Evaluationmentioning

confidence: 99%

IGPred‐HDnet: Prediction of Immunoglobulin Proteins Using Graphical Features and the Hierarchal Deep Learning‐Based Approach

Ali

Alturise

Alkhalifah

et al. 2023

Computational Intelligence and Neuroscience

Self Cite

View full text Add to dashboard Cite

Motivation. Immunoglobulin proteins (IGP) (also called antibodies) are glycoproteins that act as B-cell receptors against external or internal antigens like viruses and bacteria. IGPs play a significant role in diverse cellular processes ranging from adhesion to cell recognition. IGP identifications via the in-silico approach are faster and more cost-effective than wet-lab technological methods. Methods. In this study, we developed an intelligent theoretical deep learning framework, “IGPred-HDnet” for the discrimination of IGPs and non-IGPs. Three types of promising descriptors are feature extraction based on graphical and statistical features (FEGS), amphiphilic pseudo-amino acid composition (Amp-PseAAC), and dipeptide composition (DPC) to extract the graphical, physicochemical, and sequential features. Next, the extracted attributes are evaluated through machine learning, i.e., decision tree (DT), support vector machine (SVM), k-nearest neighbour (KNN), and hierarchical deep network (HDnet) classifiers. The proposed predictor IGPred-HDnet was trained and tested using a 10-fold cross-validation and independent test. Results and Conclusion. The success rates in terms of accuracy (ACC) and Matthew’s correlation coefficient (MCC) of IGPred-HDnet on training and independent dataset (Dtrain Dtest) are ACC = 98.00%, 99.10%, and MCC = 0.958, and 0.980 points, respectively. The empirical outcomes demonstrate that the IGPred-HDnet model efficacy on both datasets using the novel FEGS feature and HDnet algorithm achieved superior predictions to other existing computational models. We hope this research will provide great insights into the large-scale identification of IGPs and pharmaceutical companies in new drug design.

show abstract

ORI-Deep: improving the accuracy for predicting origin of replication sites by using a blend of features and long short-term memory network

Cited by 16 publications

References 58 publications

DNAPred_Prot: Identification of DNA-Binding Proteins Using Composition- and Position-Based Features

DNAPred_Prot: Identification of DNA-Binding Proteins Using Composition- and Position-Based Features

Evaluation of deep learning techniques for identification of sarcoma-causing carcinogenic mutations

IGPred‐HDnet: Prediction of Immunoglobulin Proteins Using Graphical Features and the Hierarchal Deep Learning‐Based Approach

Contact Info

Product

Resources

About