Predicting protein structural classes for low-similarity sequences by evaluating different features

Zhu, Xiaojuan; Feng, Chao-Qin; Lai, Hoang M.; Chen, Wei; Lin, Hao

doi:10.1016/j.knosys.2018.10.007

Cited by 188 publications

(92 citation statements)

References 100 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Besides, previous predictors relied heavily on the features derived from third-party tools, such as position-specific score matrix (PSSM) [23], Z-coordinate, secondary structure [24], and so on. Although these features contribute to the improvement of the predictor performance [25][26][27], their weakness cannot be ignored. On the one hand, using these third-party tool-derived features will make the predictor slow and may lead to uncontrollable failure.…”

Section: Methodsmentioning

confidence: 99%

TMP-SSurface: A Deep Learning-Based Predictor for Surface Accessibility of Transmembrane Protein Residues

et al. 2019

View full text Add to dashboard Cite

Transmembrane proteins (TMPs) play vital and diverse roles in many biological processes, such as molecular transportation and immune response. Like other proteins, many major interactions with other molecules happen in TMPs' surface area, which is important for function annotation and drug discovery. Under the condition that the structure of TMP is hard to derive from experiment and prediction, it is a practical way to predict the TMP residues' surface area, measured by the relative accessible surface area (rASA), based on computational methods. In this study, we presented a novel deep learning-based predictor TMP-SSurface for both alpha-helical and beta-barrel transmembrane proteins (α-TMP and β-TMP), where convolutional neural network (CNN), inception blocks, and CapsuleNet were combined to construct a network framework, simply accepting one-hot code and position-specific score matrix (PSSM) of protein fragment as inputs. TMP-SSurface was tested against an independent dataset achieving appreciable performance with 0.584 Pearson correlation coefficients (CC) value. As the first TMP's rASA predictor utilizing the deep neural network, our method provided a referenceable sample for the community, as well as a practical step to discover the interaction sites of TMPs based on their sequence.

show abstract

Section: Methodsmentioning

confidence: 99%

TMP-SSurface: A Deep Learning-Based Predictor for Surface Accessibility of Transmembrane Protein Residues

et al. 2019

View full text Add to dashboard Cite

show abstract

“…PSI-BLAST was used to assess the PSSM for each sequence based on sequences in the non-redundant Swiss-PROT database that share significant similarity, with three iterations and an e-value threshold of 0.0001 (Bhagwat and Aravind, 2007;Zhu et al, 2019). The raw PSSMs are n × 20 matrices; n rows indicate the query protein residues with n being the length of the protein sequence and 20 columns represent the 20 standard amino acids that may exist in the related protein sequences.…”

Section: Data Setsmentioning

confidence: 99%

Identifying Antifreeze Proteins Based on Key Evolutionary Information

Sun

Ding

Wang

et al. 2020

Front. Bioeng. Biotechnol.

View full text Add to dashboard Cite

Antifreeze proteins are important antifreeze materials that have been widely used in industry, including in cryopreservation, de-icing, and food storage applications. However, the quantity of some commercially produced antifreeze proteins is insufficient for large-scale industrial applications. Further, many antifreeze proteins have properties such as cytotoxicity, severely hindering their applications. Understanding the mechanisms underlying the protein-ice interactions and identifying novel antifreeze proteins are, therefore, urgently needed. In this study, to uncover the mechanisms underlying protein-ice interactions and provide an efficient and accurate tool for identifying antifreeze proteins, we assessed various evolutionary features based on position-specific scoring matrices (PSSMs) and evaluated their importance for discriminating of antifreeze and non-antifreeze proteins. We then parsimoniously selected seven key features with the highest importance. We found that the selected features showed opposite tendencies (regarding the conservation of certain amino acids) between antifreeze and non-antifreeze proteins. Five out of the seven features had relatively high contributions to the discrimination of antifreeze and non-antifreeze proteins, as revealed by a principal component analysis, i.e., the conservation of the replacement of Cys, Trp, and Gly in antifreeze proteins by Ala, Met, and Ala, respectively, in the related proteins, and the conservation of the replacement of Arg in non-antifreeze proteins by Ser and Arg in the related proteins. Based on the seven parsimoniously selected key features, we established a classifier using support vector machine, which outperformed the state-of-the-art tools. These results suggest that understanding evolutionary information is crucial to designing accurate automated methods for discriminating antifreeze and non-antifreeze proteins. Our classifier, therefore, is an efficient tool for annotating new proteins with antifreeze functions based on sequence information and can facilitate their application in industry.

show abstract

“…Selecting representative features is a crucial step because they directly determine prediction performance [41,42]. Seven sequence-based feature extraction methods were used: local structural entropy (LSE), NetSurfP, DisEMBL, overall amino acid composition (OAAC), dipeptide composition, PSSM profiles, and physicochemical properties.…”

Section: Feature Extractionmentioning

confidence: 99%

PredPSD: A Gradient Tree Boosting Approach for Single-Stranded and Double-Stranded DNA Binding Protein Prediction

et al. 2019

View full text Add to dashboard Cite

Interactions between proteins and DNAs play essential roles in many biological processes. DNA binding proteins can be classified into two categories. Double-stranded DNA-binding proteins (DSBs) bind to double-stranded DNA and are involved in a series of cell functions such as gene expression and regulation. Single-stranded DNA-binding proteins (SSBs) are necessary for DNA replication, recombination, and repair and are responsible for binding to the single-stranded DNA. Therefore, the effective classification of DNA-binding proteins is helpful for functional annotations of proteins. In this work, we propose PredPSD, a computational method based on sequence information that accurately predicts SSBs and DSBs. It introduces three novel feature extraction algorithms. In particular, we use the autocross-covariance (ACC) transformation to transform feature matrices into fixed-length vectors. Then, we put the optimal feature subset obtained by the minimal-redundancy-maximal-relevance criterion (mRMR) feature selection algorithm into the gradient tree boosting (GTB). In 10-fold cross-validation based on a benchmark dataset, PredPSD achieves promising performances with an AUC score of 0.956 and an accuracy of 0.912, which are better than those of existing methods. Moreover, our method has significantly improved the prediction accuracy in independent testing. The experimental results show that PredPSD can significantly recognize the binding specificity and differentiate DSBs and SSBs.

show abstract

Predicting protein structural classes for low-similarity sequences by evaluating different features

Cited by 188 publications

References 100 publications

TMP-SSurface: A Deep Learning-Based Predictor for Surface Accessibility of Transmembrane Protein Residues

TMP-SSurface: A Deep Learning-Based Predictor for Surface Accessibility of Transmembrane Protein Residues

Identifying Antifreeze Proteins Based on Key Evolutionary Information

PredPSD: A Gradient Tree Boosting Approach for Single-Stranded and Double-Stranded DNA Binding Protein Prediction

Contact Info

Product

Resources

About