Dataset Information

Using amino acid physicochemical distance transformation for fast protein remote homology detection.

ABSTRACT: Protein remote homology detection is one of the most important problems in bioinformatics. Discriminative methods such as support vector machines (SVM) have shown superior performance. However, the performance of SVM-based methods depends on the vector representations of the protein sequences. Prior works have demonstrated that sequence-order effects are relevant for discrimination, but little work has explored how to incorporate the sequence-order information along with the amino acid physicochemical properties into the prediction. In order to incorporate the sequence-order effects into the protein remote homology detection, the physicochemical distance transformation (PDT) method is proposed. Each protein sequence is converted into a series of numbers by using the physicochemical property scores in the amino acid index (AAIndex), and then the sequence is converted into a fixed length vector by PDT. The sequence-order information can be efficiently included into the feature vector with little computational cost by this approach. Finally, the feature vectors are input into a support vector machine classifier to detect the protein remote homologies. Our experiments on a well-known benchmark show the proposed method SVM-PDT achieves superior or comparable performance with current state-of-the-art methods and its computational cost is considerably superior to those of other methods. When the evolutionary information extracted from the frequency profiles is combined with the PDT method, the profile-based PDT approach can improve the performance by 3.4% and 11.4% in terms of ROC score and ROC50 score respectively. The local sequence-order information of the protein can be efficiently captured by the proposed PDT and the physicochemical properties extracted from the amino acid index are incorporated into the prediction. The physicochemical distance transformation provides a general framework, which would be a valuable tool for protein-level study.

SUBMITTER: Liu B

PROVIDER: S-EPMC3460876 | biostudies-literature | 2012

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Using amino acid physicochemical distance transformation for fast protein remote homology detection.

Liu Bin B Wang Xiaolong X Chen Qingcai Q Dong Qiwen Q Lan Xun X

PloS one 20120928 9

Protein remote homology detection is one of the most important problems in bioinformatics. Discriminative methods such as support vector machines (SVM) have shown superior performance. However, the performance of SVM-based methods depends on the vector representations of the protein sequences. Prior works have demonstrated that sequence-order effects are relevant for discrimination, but little work has explored how to incorporate the sequence-order information along with the amino acid physicoch ...[more]

PMID: 23029559

Similar Datasets

Project description:Studying protein structures and dynamics directly in the cellular environments in which they function is essential to fully understand the molecular mechanisms underlying cellular processes. Site-directed spin-labeling (SDSL)-in combination with double electron-electron resonance (DEER) spectroscopy-has emerged as a powerful technique for determining both the structural states and the conformational equilibria of biomacromolecules. In-cell DEER spectroscopy on proteins in mammalian cells has thus far not been possible due to the notable challenges of spin-labeling in live cells. In-cell SDSL requires exquisite biorthogonality, high labeling reaction rates and low background signal from unreacted residual spin label. While the bioorthogonal reaction must be highly specific and proceed under physiological conditions, many spin labels display time-dependent instability in the reducing cellular environment. Additionally, high concentrations of spin label can be toxic. Thus, an exceptionally fast bioorthogonal reaction is required that can allow for complete labeling with low concentrations of spin-label prior to loss of signal. Here we utilized genetic code expansion to site-specifically encode a novel family of small, tetrazine-bearing non-canonical amino acids (Tet-v4.0) at multiple sites in green fluorescent protein (GFP) and maltose binding protein (MBP) expressed both in E. coli and in human HEK293T cells. We achieved specific and quantitative spin-labeling of Tet-v4.0-containing proteins by developing a series of strained trans -cyclooctene (sTCO)-functionalized nitroxides-including a gem -diethyl-substituted nitroxide with enhanced stability in cells-with rate constants that can exceed 10 6 M -1 s -1 . The remarkable speed of the Tet-v4.0/sTCO reaction allowed efficient spin-labeling of proteins in live HEK293T cells within minutes, requiring only sub-micromolar concentrations of sTCO-nitroxide added directly to the culture medium. DEER recorded from intact cells revealed distance distributions in good agreement with those measured from proteins purified and labeled in vitro . Furthermore, DEER was able to resolve the maltose-dependent conformational change of Tet-v4.0-incorporated and spin-labeled MBP in vitro and successfully discerned the conformational state of MBP within HEK293T cells. We anticipate the exceptional reaction rates of this system, combined with the relatively short and rigid side chains of the resulting spin labels, will enable structure/function studies of proteins directly in cells, without any requirements for protein purification.Toc

Dataset Information

Using amino acid physicochemical distance transformation for fast protein remote homology detection.

Publications

Using amino acid physicochemical distance transformation for fast protein remote homology detection.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets