A two-stage approach towards protein secondary structure classification

Ghosh, Kushal Kanti; Ghosh, Soulib; Sen, Sagnik; Sarkar, Ram; Maulik, Ujjwal

doi:10.1007/s11517-020-02194-w

Cited by 15 publications

(7 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…We have used random forest (RF) for classification purposes. RF follows a bootstrapping algorithm (Ghosh et al 2020) depending on a decision tree model, which is very helpful in solving the over fitting problem. A smaller number of hyper-parameters and a good prediction result also make it simple and useful.…”

Section: Classifier Selectionmentioning

confidence: 99%

Application of texture-based features for text non-text classification in printed document images with novel feature selection algorithm

et al. 2021

Self Cite

View full text Add to dashboard Cite

Text non-text separation is one of the most essential pre-processing steps for any optical character recognition (OCR) system. As an OCR engine can only process texts, the non-texts present in an input document image are required to be suppressed at the initial level. Therefore, to build a complete OCR system, an efficient text non-text separation module needs to be developed. To this end, we have proposed a texture-based feature descriptor followed by a novel feature selection technique for region-based text non-text classification. First, we have incorporated rotation invariant property with local ternary pattern to form a new texture-based feature descriptor, rotation invariant local ternary pattern (RILTP). Next, a novel feature selection technique is proposed which is a modified version of binary particle swarm optimization (BPSO). For the evaluation of the proposed text non-text classification method, we have initially constructed a database consisting of 690 images of text and non-text regions extracted from 70 pages of RDCL 2015 and 75 pages of RDCL 2017 page segmentation competitions databases. In this database, each class contains 345 data samples. The proposed texturebased feature descriptor has obtained an accuracy of 97.09% on this database. Whereas, after applying BPSO, the feature dimension is reduced by approximately 55% and at the same time, the accuracy reaches 97.5%. Furthermore, in this work, another database is also created from Media team document pages to validate the robustness of this method. The second database comprises 100 text and 100 non-text images. The method has achieved 96.28% accuracy when it is trained with the first database and tested with the second database. The comparative study reveals the robustness and strength of the proposed method as it outnumbers many state-of-the-art texture-based features. Besides, the proposed feature selection method is also compared with various standard feature selection methods, and it has been observed that the proposed one outperforms all those methods considered here for comparison.

show abstract

Section: Classifier Selectionmentioning

confidence: 99%

Application of texture-based features for text non-text classification in printed document images with novel feature selection algorithm

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…Therefore, more advanced machine learning (ML) and deep learning (DL) methods have been widely used in protein structure prediction, leading to significant progress. 3,4 With the development of sequencing technology and related research, the number of protein primary structure sequences (i.e., amino acid sequence of protein composition) has exponentially increased, and many of these sequences have been stored in large biological databases. However, due to the complex arrangement and diverse structures of amino acids under the tertiary structure, it is extremely challenging to directly use the primary structure to the predict tertiary structure.…”

Section: ■ Introductionmentioning

confidence: 99%

“…However, these methods for determining the structure have disadvantages such as high cost and extended time requirements. Therefore, more advanced machine learning (ML) and deep learning (DL) methods have been widely used in protein structure prediction, leading to significant progress. , …”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

PSSP-MFFNet: A Multifeature Fusion Network for Protein Secondary Structure Prediction

Chen,

Chen

2024

ACS Omega

View full text Add to dashboard Cite

Protein secondary structure prediction (PSSP) is a fundamental task in modern bioinformatics research and is particularly important for uncovering the functional mechanisms of proteins. To improve the accuracy of PSSP, various general and essential features generated from amino acid sequences are often used for predicting the secondary structure. In this paper, we propose PSSP-MFFNet, a deep learning-based multi-feature fusion network for PSSP, which incorporates a multi-view deep learning architecture with the multiple sequence alignment (MSA) Transformer to efficiently capture global and local features of protein sequences. In practice, PSSP-MFFNet adopts a multi-feature fusion strategy, integrating different features generated from protein sequences, including MSA, sequence information, evolutionary information, and hidden state information. Moreover, we employ the MSA Transformer to interleave row and column attention across the input MSA. A hybrid network architecture of convolutional neural networks and long short-term memory networks is applied to extract high-level features after feature fusion. Furthermore, we introduce a transformer encoder to enhance the extracted high-level features. Comparative experimental results on independent tests demonstrate that PSSP-MFFNet has excellent generalization ability, outperforming other state-of-the-art PSSP models by an average of 1% on public benchmarks, including CASP12, CASP13, CASP14, TEST2018, and CB513. Our method can contribute to a better understanding of the biological functions of proteins, which has significant implications for drug discovery, disease diagnosis, and protein engineering.

show abstract

“…In 34 was attempted to show Artificial Neural Network (ANN) with different feature extraction method was more accurate than other classifier methods. In another study, Gosh et al 35 proposed a two-stage framework for feature extraction and classification. They utilized sequence-based and structure-based features in their framework which removed redundant features by, mutual information (MI) feature selection method.…”

Section: Introductionmentioning

confidence: 99%

A novel fusion based on the evolutionary features for protein fold recognition using support vector machines

Refahi

Mir

Nasiri

2020

Sci Rep

View full text Add to dashboard Cite

protein fold recognition plays a crucial role in discovering three-dimensional structure of proteins and protein functions. Several approaches have been employed for the prediction of protein folds. Some of these approaches are based on extracting features from protein sequences and using a strong classifier. Feature extraction techniques generally utilize syntactical-based information, evolutionarybased information and physicochemical-based information to extract features. In recent years, finding an efficient technique for integrating discriminate features have been received advancing attention. in this study, we integrate Auto-cross-covariance and Separated dimer evolutionary feature extraction methods. The results' features are scored by Information gain to define and select several discriminated features. According to three benchmark datasets, DD, RDD ,and eDD, the results of the support vector machine show more than 6 % improvement in accuracy on these benchmark datasets. Proteins are Jack of all trades biological macromolecules. They are involved in almost every biological reaction; Protein plays a critical role in many different areas such as building muscle, hormone production, enzyme, immune function, and energy. Typically more than 20,000 proteins exist in human cells 1 , to acquire knowledge about the protein function and interactions, the prediction of protein structural classes is extremely useful 2. Fold recognition is one of the fundamental methods in protein structure and function prediction. Each type of protein has a particular three-dimensional structure, which is determined by the order of the amino acids in its polypeptide chain. A protein' s structure begins with its amino acid sequence, which is thus considered its primary structure. The next level of the organization includes the α helix and β sheets that forms with certain segments of the polypeptide chain; these folds are elements of secondary structure. The full, threedimensional conformation formed by an entire polypeptide chain is referred to as tertiary structure 3. One of the main steps which can be assumed as a vital stage for predicting protein fold(secondary structure) is feature extraction. Computational feature extraction methods are divided into syntactical, physicochemical and evolutionary methods. Syntactical methods pay attention only to the protein sequence, like composition and occurrence 4-6. physicochemical methods consider some physical and chemical properties of protein sequences. Evolutionary methods extract features from Basic Local Alignment Search Tool(BLAST). When attempting to solve many biological problems, it is obvious that a single data source might not be informative, and combining several complementary biological data sources will lead to a more accurate result. When we studied methods of protein fold recognition, we found that less attention has been paid to the fusion of features to get more comprehensive features. In recent studies, researchers attempted to find new feature extraction methods 7-12 or t...

show abstract

A two-stage approach towards protein secondary structure classification

Cited by 15 publications

References 55 publications

Application of texture-based features for text non-text classification in printed document images with novel feature selection algorithm

Application of texture-based features for text non-text classification in printed document images with novel feature selection algorithm

PSSP-MFFNet: A Multifeature Fusion Network for Protein Secondary Structure Prediction

A novel fusion based on the evolutionary features for protein fold recognition using support vector machines

Contact Info

Product

Resources

About