Prediction of protein secondary structures with a novel kernel density estimation based classifier

Chang, Darby Tien Hao; Ou, Yu‐Yen; Hung, Hao Geng; Yang, Meng; Chen, Chien‐Yu; Oyang, Yen Jen

doi:10.1186/1756-0500-1-51

Cited by 11 publications

(5 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The execution time is measured on a workstation equipped with an Intel Core 2 Duo E8400 3.0 GHz CPU and 8 GB memory, and do not include the time taken to carry out parameter selection or cross validation. According to the observed time complexity of SVM [ 41 ], a complete parameter selection on a 1:15 dataset of the same amount of positive samples may take months or even years using a contemporary workstation. Conversely, analyzing the same 1:1 dataset with RVKDE takes only 142 seconds on the same workstation mentioned above, allowing for the analysis of unbalanced datasets within a reasonable time.…”

Section: Resultsmentioning

confidence: 99%

Predicting protein-protein interactions in unbalanced data using the primary structure of proteins

Chou

Chang

2010

BMC Bioinformatics

Self Cite

View full text Add to dashboard Cite

BackgroundElucidating protein-protein interactions (PPIs) is essential to constructing protein interaction networks and facilitating our understanding of the general principles of biological systems. Previous studies have revealed that interacting protein pairs can be predicted by their primary structure. Most of these approaches have achieved satisfactory performance on datasets comprising equal number of interacting and non-interacting protein pairs. However, this ratio is highly unbalanced in nature, and these techniques have not been comprehensively evaluated with respect to the effect of the large number of non-interacting pairs in realistic datasets. Moreover, since highly unbalanced distributions usually lead to large datasets, more efficient predictors are desired when handling such challenging tasks.ResultsThis study presents a method for PPI prediction based only on sequence information, which contributes in three aspects. First, we propose a probability-based mechanism for transforming protein sequences into feature vectors. Second, the proposed predictor is designed with an efficient classification algorithm, where the efficiency is essential for handling highly unbalanced datasets. Third, the proposed PPI predictor is assessed with several unbalanced datasets with different positive-to-negative ratios (from 1:1 to 1:15). This analysis provides solid evidence that the degree of dataset imbalance is important to PPI predictors.ConclusionsDealing with data imbalance is a key issue in PPI prediction since there are far fewer interacting protein pairs than non-interacting ones. This article provides a comprehensive study on this issue and develops a practical tool that achieves both good prediction performance and efficiency using only protein sequence information.

show abstract

Section: Resultsmentioning

confidence: 99%

Predicting protein-protein interactions in unbalanced data using the primary structure of proteins

Chou

Chang

2010

BMC Bioinformatics

Self Cite

View full text Add to dashboard Cite

show abstract

“…The fixed length format was obtained from protein sequences of variable length using amino acid and dipeptide composition. It has been successfully applied to numerous classification and pattern recognition problems such as classification of microarray data, protein secondary structure prediction and sub cellular localization [3,4,18].…”

Section: Methodsmentioning

confidence: 99%

“…The support vector machine (SVM) based prediction system is fully automatic and reliable. It has been used in many applications including sub-cellular localization, protein secondary structure prediction, and micro array data analysis of proteins [1][2][3][4]. However, no direct method is currently available to predict blood proteins.…”

Section: Introductionmentioning

confidence: 99%

Analysis and Prediction of Major Blood Proteins Based on Their Amino Acid and Dipeptide Composition

Muthukrishnan

Puri²,

Lefèvre

2013

Int J of Bioi Res

View full text Add to dashboard Cite

ResultsTo develop a prediction platform for blood proteins, amino acid composition dataset of all blood-proteins were made using five fold cross-validation technique. The SVM module was developed using blood-protein and non-blood protein training sets. Dataset was divided into five equal sets randomly, four sets were used for training, and the remaining set was used for testing [10,11]. This process was repeated five times to test each protein at least once [12]. We labeled all blood-proteins as positive proteins and non bloodproteins were used as negative proteins. The results demonstrated

show abstract

“…Berdasarkan metode yang diusulkan dari Gromiha dan Yabuki [5] dengan menggunakan neural network dalam mengklasifikasikan transport protein ke dalam tiga kelas utama yaitu Channels/pores, electrochemical, dan active transporters. Dalam penelitian sebelumnya juga, Ou telah menganalisis komposisi asam amino, komposisi pasangan residu dan sifat asam amino dalam tiga kelas dan enam family dengan menggunakan metode Position Specific Scoring Matrix (PSSM) dengan tingkat akurasi sebesar 78% [6][7][8]. Penelitian sebelumnya juga telah membuat tools untuk prediksi protein transport menggunakan PSSM [7].…”

Section: Pendahuluanunclassified

Klasifikasi Fungsi Family Protein Transport Menggunakan Radial Basis Neural Network

Sandag

Kaunang

2019

CogITo Smart Journal

View full text Add to dashboard Cite

AbstrakTransporter adalah protein transmembran yang penting dalam proses masuk dan keluarnya ion atau molekul sel di seluruh protein membran dan memainkan peran penting dalam mengenali sistem kekebalan tubuh dan transduser energi. Dalam beberapa tahun terakhir, penelitian sebelumnya telah dilakukan untuk menganalisis protein transport, terutama diskriminasi kelas dan familynya dalam memainkan peran penting dalam system control sel, mengangkut air, sinyal kimia dan listrik. Protein transport membrane cenderung membentuk system pompa dan channel span, serta span cell membrane. Oleh karena itu, membedakan kelas dan family transport protein adalah tugas penting dalam ilmu komputasi biologi dan diperlukan bagi para ahli biologi untuk mendapatkan pemahaman yang lebih baik tentang fungsi protein transport. Oleh karena itu, dalam penelitian ini, telah dilakukan pengembangan metode untuk mengidentifikasi fungsi kelas utama dan family protein transport menggunakan radial basis neural network. Peneliti telah mengalanisis karakteristik komposisi asam amino, komposisi residu pair pada protein transport. Metode dalam klasifikasi kelas protein transport untuk mengetahui fungsi protein transport peneliti menggunakan PSSM dengan metode quickRBF classifier memberikan hasil akurasi terbaik dibanding dengan metode yang lain. Hasil akurasi sebesar 84,84% untuk cross validation dan 80,71% untuk independent data, oleh karena itu maka motode yang peneliti usulkan dapat digunakan secara efektif untuk mengidentifikasi dan mendiskriminasi transporter ke dalam kelas protein transport dengan peningkatan 6-10 % dari penelitian yang sejenis. AbstractTransporters are important transmembrane proteins that involve the cellular entry and exit of ions or molecules throughout the membrane proteins and thereby play important roles in recognizing the immune system and energy transducers. In recent years, several studies have been conducted to analyze the transport proteins; especially the discrimination class of transporters and their subfamilies play crucial roles in cell control system, transporting water, chemical and electric signals. Membrane transport proteins tend to form an intricate system of pumps and channel span, and span cell membranes. Hence, discriminating the specific class of transporters and their subfamilies is an essential task in computational biology and necessary for biologists to gain better understanding about the function of transport proteins. Therefore, in this study, an attempt has been made to develop a method that used radial basis neural network to identify the function of transport proteins in major class and family. We have Cogito Smart Journal | VOL. 5 | NO.2 | Desember 2019 n 204Fakultas Ilmu Komputer | Universitas Klabat | CORIS | analyzed the charateristics of amino acid composition (AAC), dipeptide pair composition (DPC) in transport proteins, also we used PSSM with quickRBF classifier method give the best accuracy results compared to other methods. Accuracy results of 84.84% for cross validation and 80.71% for i...

show abstract

Prediction of protein secondary structures with a novel kernel density estimation based classifier

Cited by 11 publications

References 14 publications

Predicting protein-protein interactions in unbalanced data using the primary structure of proteins

Predicting protein-protein interactions in unbalanced data using the primary structure of proteins

Analysis and Prediction of Major Blood Proteins Based on Their Amino Acid and Dipeptide Composition

Klasifikasi Fungsi Family Protein Transport Menggunakan Radial Basis Neural Network

Contact Info

Product

Resources

About