SDBP-Pred: Prediction of single-stranded and double-stranded DNA-binding proteins by extending consensus sequence and K-segmentation strategies into PSSM

Ali, Farman; Arif, Muhammad; Khan, Zaheer; Kabir, Muhammad; Ahmad, Saeed; Yu, Dong‐Jun

doi:10.1016/j.ab.2019.113494

Cited by 35 publications

(21 citation statements)

References 65 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Compared to the large number of publications on prediction of DNA binding proteins, the investigation on ssDNA binding protein prediction is limited so far. To our knowledge, currently there are only four published studies on SSB prediction using machine learning-based approaches [ 95 , 96 , 97 , 98 ]. These methods typically consist of four major steps as shown in Figure 3 : (1) dataset generation for training and testing; (2) features for learning and prediction; (3) classification models; and (4) performance evaluation.…”

Section: Machine Learning-based Methods For Ssb Predictionmentioning

confidence: 99%

“…So far, the prediction of SSBs has been generally carried out for differentiation between SSBs and dsDNA binding proteins (DSBs). Essentially all the methods use the same training and independent test sets from the work by Wang et al [ 95 , 96 , 97 , 98 ]. The training set (Uniprot1065) consists of 873 DSBs and 183 SSBs, which were culled from UniProtKB/SwissProt with a sequence identity cutoff at 0.7.…”

Section: Machine Learning-based Methods For Ssb Predictionmentioning

confidence: 99%

“…It is understandable that a higher sequence identity cutoff was adopted for the training dataset construction due to limited availability of annotated SSBs. However, it may cause an overfitting problem in machine learning models as evidenced by the unbalanced performance between the training set and the independent dataset, resulting in a big drop of MCC (Matthew’s correlation coefficient) from the training set to the independent set, regardless of the type of models or features used for the prediction [ 95 , 96 , 97 , 98 ]. Another potential issue that may contribute to the unbalanced performance is the phenomenon of protein moonlighting since some DNA binding proteins can bind both ssDNA and dsDNA.…”

Section: Machine Learning-based Methods For Ssb Predictionmentioning

confidence: 99%

See 2 more Smart Citations

Single-Stranded DNA Binding Proteins and Their Identification Using Machine Learning-Based Approaches

Guo

Malik

2022

Biomolecules

View full text Add to dashboard Cite

Single-stranded DNA (ssDNA) binding proteins (SSBs) are critical in maintaining genome stability by protecting the transient existence of ssDNA from damage during essential biological processes, such as DNA replication and gene transcription. The single-stranded region of telomeres also requires protection by ssDNA binding proteins from being attacked in case it is wrongly recognized as an anomaly. In addition to their critical roles in genome stability and integrity, it has been demonstrated that ssDNA and SSB–ssDNA interactions play critical roles in transcriptional regulation in all three domains of life and viruses. In this review, we present our current knowledge of the structure and function of SSBs and the structural features for SSB binding specificity. We then discuss the machine learning-based approaches that have been developed for the prediction of SSBs from double-stranded DNA (dsDNA) binding proteins (DSBs).

show abstract

Section: Machine Learning-based Methods For Ssb Predictionmentioning

confidence: 99%

Section: Machine Learning-based Methods For Ssb Predictionmentioning

confidence: 99%

Section: Machine Learning-based Methods For Ssb Predictionmentioning

confidence: 99%

See 1 more Smart Citation

Single-Stranded DNA Binding Proteins and Their Identification Using Machine Learning-Based Approaches

Guo

Malik

2022

Biomolecules

View full text Add to dashboard Cite

show abstract

“…All the above-cited methods have shown great contribution in prediction of drug-target interaction, however, each predictor has its limitation. For example, structure-based methods are expensive and limited applications due to the unavailability of structural information of all proteins in the databanks 19 – 21 . Most existing predictors have used conventional feature extraction methods such as amino acid composition, dipeptide composition, and position specific scoring matrix, however, these approaches do not effectively explore the important features.…”

Section: Introductionmentioning

confidence: 99%

XGB-DrugPred: computational prediction of druggable proteins using eXtreme gradient boosting and optimized features set

Sikander

Ghulam

Ali

2022

Sci Rep

Self Cite

View full text Add to dashboard Cite

Accurate identification of drug-targets in human body has great significance for designing novel drugs. Compared with traditional experimental methods, prediction of drug-targets via machine learning algorithms has enhanced the attention of many researchers due to fast and accurate prediction. In this study, we propose a machine learning-based method, namely XGB-DrugPred for accurate prediction of druggable proteins. The features from primary protein sequences are extracted by group dipeptide composition, reduced amino acid alphabet, and novel encoder pseudo amino acid composition segmentation. To select the best feature set, eXtreme Gradient Boosting-recursive feature elimination is implemented. The best feature set is provided to eXtreme Gradient Boosting (XGB), Random Forest, and Extremely Randomized Tree classifiers for model training and prediction. The performance of these classifiers is evaluated by tenfold cross-validation. The empirical results show that XGB-based predictor achieves the best results compared with other classifiers and existing methods in the literature.

show abstract

“…It requires a sufficient quantity of cells, therefore, human male semen cells are usually used. From those cells, one can either observe locations where the chromosome broke and recombination occurred using, for instance, fluorescence [24], illumina sequencing [25], or identifying binding proteins [26] and their binding sites [27][28][29].…”

mentioning

confidence: 99%

High fraction of silent recombination in a finite population two-locus neutral birth-death-mutation model

Melka,

Louzoun

2021

Preprint

View full text Add to dashboard Cite

A precise estimate of allele and haplotype polymorphism is of great interest for theoretical population genetics, but also practical issues, such as bone marrow registries. Allele polymorphism is driven mainly by point mutations, while haplotype polymorphism is also affected by recombination events. Even in the simple case of two loci in a haploid individual, there is currently no good estimate of the number of haplotypes as a function of the mutation and recombination rates. We here propose such an estimate and show that the common approximation that recombination can be treated as mutations is limited to recombination rates of the same order as the mutation rate. Beyond this regime, the total number of haplotypes is much lower than expected from the approximation above. Moreover, in contrast with mutations, the number of haplotypes does not grow linearly with the population size. We apply this new estimate to very large-scale human haplotype frequencies from human populations to show that the current estimated haplotype recombination rate in the HLA region is underestimated. This high recombination rate may be the source of HLA haplotype extreme polymorphism.

show abstract

SDBP-Pred: Prediction of single-stranded and double-stranded DNA-binding proteins by extending consensus sequence and K-segmentation strategies into PSSM

Cited by 35 publications

References 65 publications

Single-Stranded DNA Binding Proteins and Their Identification Using Machine Learning-Based Approaches

Single-Stranded DNA Binding Proteins and Their Identification Using Machine Learning-Based Approaches

XGB-DrugPred: computational prediction of druggable proteins using eXtreme gradient boosting and optimized features set

High fraction of silent recombination in a finite population two-locus neutral birth-death-mutation model

Contact Info

Product

Resources

About