Protein secondary structure: category assignment and predictability

Andersen, Claus A.; Bohr, Henrik; Brunak, Søren

doi:10.1016/s0014-5793(01)02910-6

Cited by 24 publications

(3 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The architecture of our DeepCNF model is mainly determined by the following 3 factors (see Figure 2): (i) the number of hidden layers; (ii) the number of different neurons at each layer; and (iii) the window size at each layer. We fix the window size to 11 because the average length of an alpha helix is around eleven residues 58 and that of a beta strand is around six 59 . To show the relationship between the performance and the number of hidden layers, we trained four different DeepCNF models with 1, 3, 5, and 7 layers, respectively.…”

Section: Datasetmentioning

confidence: 99%

Protein Secondary Structure Prediction Using Deep Convolutional Neural Fields

Wang

Peng

et al. 2016

Sci Rep

491

499

View full text Add to dashboard Cite

Protein secondary structure (SS) prediction is important for studying protein structure and function. When only the sequence (profile) information is used as input feature, currently the best predictors can obtain ~80% Q3 accuracy, which has not been improved in the past decade. Here we present DeepCNF (Deep Convolutional Neural Fields) for protein SS prediction. DeepCNF is a Deep Learning extension of Conditional Neural Fields (CNF), which is an integration of Conditional Random Fields (CRF) and shallow neural networks. DeepCNF can model not only complex sequence-structure relationship by a deep hierarchical architecture, but also interdependency between adjacent SS labels, so it is much more powerful than CNF. Experimental results show that DeepCNF can obtain ~84% Q3 accuracy, ~85% SOV score, and ~72% Q8 accuracy, respectively, on the CASP and CAMEO test proteins, greatly outperforming currently popular predictors. As a general framework, DeepCNF can be used to predict other protein structure properties such as contact number, disorder regions, and solvent accessibility.

show abstract

Section: Datasetmentioning

confidence: 99%

Protein Secondary Structure Prediction Using Deep Convolutional Neural Fields

Wang

Peng

et al. 2016

Sci Rep

491

499

View full text Add to dashboard Cite

show abstract

“…For the CNN-based prediction model, we use five CNN layers (Wang et al, 2016). Since the average length of an alpha helix is around 11 residues (Andersen et al, 2001) and that of a beta strand is around 6 (Penel et al, 2003), we fix the window size to 11. For the LSTM-based networks, we apply two stacked bidirectional LSTM neural networks (Sønderby and Winther, 2014) and a FC layer.…”

Section: Prediction Networkmentioning

confidence: 99%

Comprehensive Study on Enhancing Low-Quality Position-Specific Scoring Matrix with Deep Learning for Accurate Protein Structure Property Prediction: Using Bagging Multiple Sequence Alignment Learning

Guo

et al. 2021

Journal of Computational Biology

View full text Add to dashboard Cite

Accurate predictions of protein structure properties, for example, secondary structure and solvent accessibility, are essential in analyzing the structure and function of a protein. Position-specific scoring matrix (PSSM) features are widely used in the structure property prediction. However, some proteins may have low-quality PSSM features due to insufficient homologous sequences, leading to limited prediction accuracy. To address this limitation, we propose an enhancing scheme for PSSM features. We introduce the ''Bagging MSA'' (multiple sequence alignment) method to calculate PSSM features used to train our model, adopt a convolutional network to capture local context features and bidirectional long shortterm memory for long-term dependencies, and integrate them under an unsupervised framework. Structure property prediction models are then built upon such enhanced PSSM features for more accurate predictions. Moreover, we develop two frameworks to evaluate the effectiveness of the enhanced PSSM features, which also bring proposed method into real-world scenarios. Empirical evaluation of CB513, CASP11, and CASP12 data sets indicates that our unsupervised enhancing scheme indeed generates more informative PSSM features for structure property prediction.

show abstract

“…It plays an important role in constructing sequence4 and structure5 alignments. Accurate and consistent assignment of secondary structure defines the quality of a training set for secondary structure prediction algorithms, and it is necessary to develop such algorithms with maximum predictive power 6…”

Section: Introductionmentioning

confidence: 99%

New method for protein secondary structure assignment based on a simple topological descriptor

et al. 2005

View full text Add to dashboard Cite

A simple, five-element descriptor, derived from the Delaunay tessellation of a protein structure in a single point per residue representation, can be assigned to each residue in the protein. The descriptor characterizes main-chain topology and connectivity in the neighborhood of the residue and does not explicitly depend on putative hydrogen bonds or any geometric parameter, including bond length, angles, and areas. Rules based on this descriptor can be used for accurate, robust, and computationally efficient secondary structure assignment that correlates well with the existing methods.

show abstract

Protein secondary structure: category assignment and predictability

Cited by 24 publications

References 19 publications

Protein Secondary Structure Prediction Using Deep Convolutional Neural Fields

Protein Secondary Structure Prediction Using Deep Convolutional Neural Fields

Comprehensive Study on Enhancing Low-Quality Position-Specific Scoring Matrix with Deep Learning for Accurate Protein Structure Property Prediction: Using Bagging Multiple Sequence Alignment Learning

New method for protein secondary structure assignment based on a simple topological descriptor

Contact Info

Product

Resources

About