Folding rate prediction based on neural network model

Zhang, Linxi; Li, Jing; Jiang, Zhouting; Xia, Agen

doi:10.1016/s0032-3861(03)00021-1

Cited by 38 publications

(33 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The first descriptor used in this study was TCD [21], which includes sequence distance per contact and total number of contacts per residue simultaneously. A larger sequence distance between two residues means a greater physical distance in the coil state and a greater physical distance will take a longer time for the two residues to make contact.…”

Section: Discussion Of the Descriptorsmentioning

confidence: 99%

“…The Protein Data Bank (PDB) codes of 28 two-state proteins and their logarithms of folding rates (expressed as ln k f ) are listed in Table 1 [21]. All the amino acid sequences and 3D structures of proteins used for descriptors calculation are taken from PDB [31].…”

Section: Data Setmentioning

confidence: 99%

“…Gromiha [20] used amino acid properties combined with a multiple regression method to study the relationship between the structural properties of amino acid combined and the folding rates of proteins. Zhang et al [21] devised a method of Back-Propagation Neural Network (BPNN) to build a predictive model based on three topological descriptors. However, in the previous works, the deviation between the experimental and the predicted ln k f is large for some proteins.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Accurate Prediction of the Folding Rate for Two‐State Proteins Based on Amino Acid Sequences

Zhao

Zhang

et al. 2007

QSAR Comb. Sci.

View full text Add to dashboard Cite

One of the most important challenges in biology is to understand and simulate the folding behavior of some simple two-state proteins. In this work, a large pool for descriptors of proteins was developed including amino acid descriptors, topological descriptors and charged partial surface area descriptors. A heuristic method was employed to select significant features. As a result, three descriptors, total contact distance, unfolding entropy change and total charge weighted partial negative surface area were chosen. Total contact distance was proved to be the most relevant factor controlling the folding behavior. Based on these descriptors, a support vector machine method was applied to build a predictive model. Comparing the statistical results of other works and methods, support vector machine method exhibited the best whole performance. Our results demonstrate that the native-state topology is the major determinant for the folding rates of two-state proteins and that the support vector machine method is a powerful tool to build a predictive model. In turn, the method and the knowledge can be used to develop in-silico predictive models to simulate the folding process of proteins.

show abstract

Section: Discussion Of the Descriptorsmentioning

confidence: 99%

Section: Data Setmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Accurate Prediction of the Folding Rate for Two‐State Proteins Based on Amino Acid Sequences

Zhao

Zhang

et al. 2007

QSAR Comb. Sci.

View full text Add to dashboard Cite

show abstract

“…However, the precise relationship between these characteristics and the rate are unknown. A backpropagation neural network was used to quantify this relationship [37]. Their results showed that correlations exist between these properties and the folding rate with relative errors for predicted results lower than competing methods.…”

Section: Folding Rate Predictionmentioning

confidence: 99%

Adaptive local learning in sampling based motion planning for protein folding

Ekenna

Thomas

Amato

2015

2015 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

View full text Add to dashboard Cite

Background: Simulating protein folding motions is an important problem in computational biology. Motion planning algorithms, such as Probabilistic Roadmap Methods, have been successful in modeling the folding landscape. Probabilistic Roadmap Methods and variants contain several phases (i.e., sampling, connection, and path extraction). Most of the time is spent in the connection phase and selecting which variant to employ is a difficult task. Global machine learning has been applied to the connection phase but is inefficient in situations with varying topology, such as those typical of folding landscapes. Results: We develop a local learning algorithm that exploits the past performance of methods within the neighborhood of the current connection attempts as a basis for learning. It is sensitive not only to different types of landscapes but also to differing regions in the landscape itself, removing the need to explicitly partition the landscape. We perform experiments on 23 proteins of varying secondary structure makeup with 52-114 residues. We compare the success rate when using our methods and other methods. We demonstrate a clear need for learning (i.e., only learning methods were able to validate against all available experimental data) and show that local learning is superior to global learning producing, in many cases, significantly higher quality results than the other methods. Conclusions:We present an algorithm that uses local learning to select appropriate connection methods in the context of roadmap construction for protein folding. Our method removes the burden of deciding which method to use, leverages the strengths of the individual input methods, and it is extendable to include other future connection methods.

show abstract

“…Baker's group made an important observation in 1998 that the folding rates of twostate folding proteins correlate with the native topologies and proposed a concept of contact order (CO) (Plaxco et al 1998a) to predict the protein folding rates. Since then, a great deal of studies Fiebig and Dill 1993;Plaxco and Baker 1998b;Alm and Baker 1999;Debe and Goddard 1999;Mounoz and Eaton 1999;Dinner and Karplus 2001;Gromiha and Selvaraj 2001;Mirny and Shakhnovich 2001;Zhou and Zhou 2002;Gong et al 2003;Ivankov et al 2003;Nölting et al 2003;Zhang et al 2003;Ivankov and Finkelstein 2004) has shown that the protein folding rates correlated significantly with protein's three-dimensional or secondary structures. However, these conclusions are all based on the knowledge of the native structure of proteins.…”

Section: Introductionmentioning

confidence: 99%

The influence of protein coding sequences on protein folding rates of all-β proteins

Li¹,

Liang²

2011

gpb

View full text Add to dashboard Cite

Abstract. It is currently believed that the protein folding rate is related to the protein structures and its amino acid sequence. However, few studies have been done on the problem that whether the protein folding rate is influenced by its corresponding mRNA sequence. In this paper, we analyzed the possible relationship between the protein folding rates and the corresponding mRNA sequences. The content of guanine and cytosine (GC content) of palindromes in protein coding sequence was introduced as a new parameter and added in the Gromiha's model of predicting protein folding rates to inspect its effect in protein folding process. The multiple linear regression analysis and jackknife test show that the new parameter is significant. The linear correlation coefficient between the experimental and the predicted values of the protein folding rates increased significantly from 0.96 to 0.99, and the population variance decreased from 0.50 to 0.24 compared with Gromiha's results. The results show that the GC content of palindromes in the corresponding protein coding sequence really influences the protein folding rate. Further analysis indicates that this kind of effect mostly comes from the synonymous codon usage and from the information of palindrome structure itself, but not from the translation information from codons to amino acids.Key words: All-β protein -Protein folding rate -Protein coding sequence -GC content of palindromes -Synonymous codon usage Abbreviations: GC content, guanine and cytosine content of protein coding sequence; P GC , GC content of palindromes; C GC , GC content of protein coding sequence.

show abstract

Folding rate prediction based on neural network model

Cited by 38 publications

References 35 publications

Accurate Prediction of the Folding Rate for Two‐State Proteins Based on Amino Acid Sequences

Accurate Prediction of the Folding Rate for Two‐State Proteins Based on Amino Acid Sequences

Adaptive local learning in sampling based motion planning for protein folding

The influence of protein coding sequences on protein folding rates of all-β proteins

Contact Info

Product

Resources

About