检索结果-内蒙古大学图书馆

Combining discriminant models with new multi-class SVMs

PATTERN ANALYSIS AND APPLICATIONS 2002年第2期5卷 168-179页

作者： Guermeur, Y LORIA F-54506 Vandoeuvre Les Nancy France

The idea of performing model combination, instead of model selection, has a long theoretical background in statistics. However, making use of theoretical results is ordinarily subject to the satisfaction of strong hypotheses (weak error correlation, availability of large training sets, possibility to rerun the training procedure an arbitrary number of times, etc.). In contrast, the practitioner is frequently faced with the problem of combining a given set of pre-trained classifiers, with highly correlated errors, using only a small training sample. Overfitting is then the main risk, which cannot be overcome but with a strict complexity control of the combiner selected. This suggests that SVMs should be well suited for these difficult situations. Investigating this idea, we introduce a family of multi-class SVMs and assess them as ensemble methods on a real-world problem. This task, protein secondary structure prediction, is an open problem in biocomputing for which model combination appears to be an issue of central importance. Experimental evidence highlights the gain in quality resulting from combining some of the most widely used prediction methods with our SVMs rather than with the ensemble methods traditionally used in the field. The gain increases when the outputs of the combiners are post-processed with a DP algorithm.

关键词： classifier fusion generalisation performance hierarchical sequence processing systems protein secondary structure prediction statistical learning theory Support Vector Machines

来源：评论

学校读者我要写书评

暂无评论

Special HMW-GSs and their genes of Triticum turgidum subsp dicoccoides accession D141 and the potential utilization in common wheat

引用

GENETIC RESOURCES AND CROP EVOLUTION 2016年第5期63卷 833-844页

作者： Guo, Xiao-Hui Hu, Ji-Liang Wu, Bi-Hua Wang, Zhen-Zhen Wang, Dong Liu, Deng-Cai Zheng, You-Liang Sichuan Agr Univ Triticeae Res Inst Chengdu 611130 Peoples R China Xuchang Univ Coll Food & Biol Engn Xuchang 461000 Peoples R China

High molecular weight glutenin subunits (HMW-GSs) are the critical components of wheat seed storage proteins. They play a crucial role in determining dough viscoelastic properties and end-use quality of wheat flour. To investigate the potential utilization of the special HMW-GSs from wild emmer wheat (2n = 4x = 28, AABB) accession D141 in common wheat (2n = 6x = 42, AABBDD), continued to our previous report of the 1Ay and its coding sequence, the other three HMW-GSs, 1Ax, 1Bx13.1 and 1By16.1, and their coding genes were isolated, characterized and transferred from D141. The 1Ax gene was silent because of containing a premature stop codon TGA at position 835-837. The ORFs of 1Bx13.1 and 1By16.1 were 2373 and 2157 bp in length, and encoded 789 and 717 amino acid residues, respectively. Compared with the other 20 Glu-1Bx alleles in GenBank, seven SNPs and one 18-bp deletion were detected in 1Bx13.1. Compared with the other 10 Glu-1By alleles in GenBank, nine SNPs were observed in 1By16.1. The mobility of 1Bx13.1 and 1By16.1 on the SDS-PAGE gel was most similar to that of 1Bx13 and 1By16 from common wheat (T. aestivum, 2n = 6x = 42, AABBDD) cv. 'Jimai 20', respectively. protein secondary structure prediction showed that the combination of 1Bx13.1 + 1By16.1 possessed more alpha-helixes with more amino acid residues than that of 1Bx13 + 1By16. It suggested that 1Bx13.1 + 1By16.1 from wild emmer accession D141 might have more positive effects on bread-making quality than the superior subunit combination 1Bx13 + 1By16. The HMW-GS composition analysis of F-1 and F-2 between common wheat for female and wild emmer accession D141 for male, indicated that the special HMW-GSs from wild emmer accession D141 could be easily introduced into common wheat (2n = 42) by general inter-species hybridization.

关键词： Gene cloning High molecular weight glutenin subunit Interspecies hybridization protein secondary structure prediction Triticum aestivum Wild emmer

来源：评论

学校读者我要写书评

暂无评论

Different methods, techniques and their limitations in protein structure prediction: A review

引用

PROGRESS IN BIOPHYSICS & MOLECULAR BIOLOGY 2022年 173卷 72-82页

作者： Bongirwar, Vrushali Mokhade, A. S. Shri Ramdeobaba Coll Engn & Management Nagpur India Visvesvaraya Natl Inst Technol Nagpur India

Because of the increase in different types of diseases in human habitats, demands for designing various types of drugs are also increasing. protein and its structure play a very important role in drug design. Therefore re-searchers from different areas like mathematics, medicines, and computer science are teaming up for getting better solutions in the said field. In this paper, we have discussed different methods of secondary and tertiary protein structure prediction (PSP), along with the limitations of different approaches. Different types of datasets used in PSP are also discussed here. This paper also tells about different performance measures to evaluate the prediction accuracy of PSP methods. Different software's/servers are available for download, which are used to find the protein structures for the input protein sequence. These softwares will also help to compare the per-formance of any new algorithm with other available methods. Details of those softwares are also mentioned in this paper.

关键词： Conformational sampling protein secondary structure prediction protein tertiary structure prediction

来源：评论

学校读者我要写书评

暂无评论

BETA-TURNS AND THEIR DISTORTIONS - A PROPOSED NEW NOMENCLATURE

引用

protein ENGINEERING 1990年第6期3卷 479-493页

作者： WILMOT, CM THORNTON, JM UNIV LONDON BIRKBECK COLL DEPT CRYSTALLOG MOLEC BIOL LAB LONDON WC1E 7HX ENGLAND

Using the new version of a protein Structural Database, BIPED, .beta.-turns have been extracted from 58 non-identical proteins (resolution .ltoreq. 2 .ANG.) using the standard criteria that the distance between ***. and Ci+***. is < 7 .ANG. and that the central residues are not helical. It has been shown that 42% of these do not fit into the eight conventional turn types (I, I'', II, II'', VIa, VIb, and VIII) as defined by .vphi.,.psi. limits for residues i + 1 and i + 2. Most of the unclassified turns are ''distortions'' of the standard turn types, lying just outside the specific limits. Eleven per cent of the turns are not related to the standard types, and have completely different .vphi.,.psi. combinations. Therefore, the conformational space available to a trans tetrapeptide has been explored to find all two-residue .vphi.,.psi. combinations which satisfy the criteria for .beta.-turn formation. The conformations generated by the search are the conformations observed in the data. On the basis of these observations a new nomenclature is suggested for .beta.-turns based on a shorthand descriptor of populated regions of the .vphi.,.psi. plot. For example, it is proposed that a type I turn should now be described as an .alpha..alpha. turn, the descriptor indicating the .vphi.,.psi. regions of the i + 1 and i + 2 positions of the turn. The use of descriptors, which convey information about the turn type conformation, should aid protein structural workers in turn classification and visualization. The sequence preferences for the .alpha..beta. turn have been elucidated: Pro, Asp and Ser at i + 1;Asn and His at i + 2;Pro at i + 3. These preferences have been explained in terms of specific interactions involving the side-chains. The .beta.-turn prediction program, BTURN 1.0, has been further developed by alterations to the calculation of parameters, the removal of incompatible multiple turn predictions and the inclusion of the distorted types .alpha..alpha. and

关键词： β-turns protein secondary structure prediction

来源：评论

学校读者我要写书评

暂无评论

A novel improved prediction of protein structural class using deep recurrent neural network

引用

EVOLUTIONARY INTELLIGENCE 2021年第2期14卷 253-260页

作者： Panda, Bishnupriya Majhi, Babita Siksha O Anusandhan Univ Inst Tech Educ & Res Dept Comp Sci & Engn Bhubaneswar Orissa India Guru Ghashidas Vishwavidyalaya Dept Comp Sci & Informat Technol Bilaspur Chhattisgarh India

For last few decades, sequence arrangement of amino acids have been utilized for the prediction of protein secondary structure. Recent methods have applied high dimensional natural language based features in machine learning models. Performance measures of machine learning based models are significantly affected by data size and data dimensionality. It is a huge challenge to develop a generic model which can be trained to perform both for small and large sized datasets in a low dimensional framework. In the present research, we suggest a low dimensional representation for both small and large sized datasets. A hybrid space of Atchley's factors II, IV, V, electron ion interaction potential and SkipGram based word2vec have been employed for amino acid sequence representation. Subsequently Stockwell transformation is applied to the representation to preserve features both in time and frequency domains. Finally, deep gated recurrent network with dropout, categorical-cross entropy error estimation and Adam optimization is used for classification purpose. The introduced method results in better prediction accuracies for both small (204,277, and 498) and large sized (PDB25, protein 640 and FC699) bench mark data sets of low sequence similarity (25-40%). The obtained classification accuracies for PDB25, 640, FC699, 498, 277, 204 datasets are 84.2%, 94.31%, 93.1%, 95.9%, 94.5% and 85.36% respectively. The major contributions in this research is that, for the first time, we verify the protein secondary structural class prediction in a very low dimensional (18-D) feature space with a novel feature representation method. Secondly, we also verify for the first time, the behaviour of deep networks for low dimensional small sized data sets.

关键词： protein secondary structure prediction SkipGram model T-SNE Atchley's factors DeepGRU Bioinformatics

来源：评论

学校读者我要写书评

暂无评论

Automatic amide I frequency selection for rapid quantification of protein secondary structure from Fourier transform infrared spectra of proteins

引用

PROTEOMICS 2002年第7期2卷 839-849页

作者： Hering, JA Innocent, PR Haris, PI De Montfort Univ Dept Biol Sci Leicester LE1 9BH Leics England De Montfort Univ Dept Comp Sci Leicester LE1 9BH Leics England

Here we report the development of a new neural network based approach for rapid quantification of protein secondary structure from Fourier transform infrared (FTIR) spectra of proteins. A technique for efficiently reducing the amount of spectral data by almost 90% is suggested to facilitate faster neural network analysis. Additionally, an automatic procedure is introduced for selecting only those regions within the amide 1 band of protein FTIR spectra, which can be best related to secondary structure contents by subsequent neural network analysis. Based on a given reference set of FTIR spectra from proteins with known secondary structure, a subset of merely 29 out of 101 amide I absorbance values could be identified, which lead to an improved prediction accuracy. The average prediction accuracy achieved for helix, sheet, turn, bend, and other is 4.96% which is better than that achieved by alternative methods that have been previously reported indicating the significant potential of this approach. Our suggested automatic amide I frequency selection procedure may be easily extended to identify promising regions from spectral data recorded by other spectroscopic techniques, like for example circular dichroism spectroscopy.

关键词： Fourier transform infrared spectroscopy genetic algorithms neural networks protein secondary structure prediction

来源：评论

学校读者我要写书评

暂无评论

Investigation of diabetes-induced effect on apex of rat heart myocardium by using cluster analysis and neural network approach: An FTIR study

引用

SPECTROSCOPY-AN INTERNATIONAL JOURNAL 2007年第5-6期21卷 269-278页

作者： Toyran, Neslihan Severcan, Feride Severcan, Mete Turan, Belma Baskent Univ Fac Med Dept Physiol TR-06530 Ankara Turkey Middle E Tech Univ Dept Biol Sci TR-06531 Ankara Turkey Middle E Tech Univ Dept Elect & Elect Engn TR-06531 Ankara Turkey Ankara Univ Sch Med Dept Biophys TR-06100 Ankara Turkey

Diabetes mellitus (DM) is a progressive chronic disorder, which affects people belonging to all age groups of the population. This disease is accompanied by a greatly increased risk of cardiovascular death. In the present study, the effects of streptozotocin (STZ)-induced type 1 diabetes on apex myocardium of the rat heart have been investigated using Fourier Transform Infrared (FTIR) spectra. The cluster analysis has been applied to FTIR spectra to differentiate the diabetic samples from the normal controls. In addition, the protein secondary structures of diabetic and normal tissues were predicted by neural networks based on the amide I band of the FTIR spectra. The findings mainly suggest that 5 weeks of diabetes alters the lipid and protein profile of normal rat heart apex myocardium, which might have an important role in understanding the molecular mechanism of diabetes-related heart diseases.

关键词： diabetes FTIR cluster analysis neural networks protein secondary structure prediction apex myocardium

来源：评论

学校读者我要写书评

暂无评论

Distributions of amino acids suggest that certain residue types more effectively determine protein secondary structure

引用

JOURNAL OF MOLECULAR MODELING 2013年第10期19卷 4337-4348页

作者： Saraswathi, S. Fernandez-Martinez, J. L. Kolinski, A. Jernigan, R. L. Kloczkowski, A. Nationwide Childrens Hosp Res Inst Battelle Ctr Math Med Columbus OH USA Univ Oviedo Dept Math Oviedo Spain Warsaw Univ Lab Theory Biopolymers Fac Chem PL-02093 Warsaw Poland Iowa State Univ Dept Biochem Biophys & Mol Biol Ames IA USA Ohio State Univ Coll Med Dept Pediat Columbus OH 43210 USA

Exponential growth in the number of available protein sequences is unmatched by the slower growth in the number of structures. As a result, the development of efficient and fast protein secondary structure prediction methods is essential for the broad comprehension of protein structures. Computational methods that can efficiently determine secondary structure can in turn facilitate protein tertiary structure prediction, since most methods rely initially on secondary structure predictions. Recently, we have developed a fast learning optimized prediction methodology (FLOPRED) for predicting protein secondary structure (Saraswathi et al. in JMM 18:4275, 2012). Data are generated by using knowledge-based potentials combined with structure information from the CATH database. A neural network-based extreme learning machine (ELM) and advanced particle swarm optimization (PSO) are used with this data to obtain better and faster convergence to more accurate secondary structure predicted results. A five-fold cross-validated testing accuracy of 83.8 % and a segment overlap (SOV) score of 78.3 % are obtained in this study. secondary structure predictions and their accuracy are usually presented for three secondary structure elements: alpha-helix, beta-strand and coil but rarely have the results been analyzed with respect to their constituent amino acids. In this paper, we use the results obtained with FLOPRED to provide detailed behaviors for different amino acid types in the secondary structure prediction. We investigate the influence of the composition, physico-chemical properties and position specific occurrence preferences of amino acids within secondary structure elements. In addition, we identify the correlation between these properties and prediction accuracy. The present detailed results suggest several important ways that secondary structure predictions can be improved in the future that might lead to improved protein design and engineering.

关键词： Amino acids Knowledge-based potentials Machine learning Neural networks Particle swarm optimization protein secondary structure prediction

来源：评论

学校读者我要写书评

暂无评论

AN ARTIFICIAL CRYSTALLINE DDT-BINDING POLYPEPTIDE

引用

FEBS LETTERS 1983年第2期157卷 247-251页

作者： MOSER, R THOMAS, RM GUTTE, B Biochemisches Institut der Universität Zürich Zürichbergstrasse 4 CH-8028 Zürich Switzerland

A hydrophobic 24-residue polypeptide that could potentially form a four-stranded antiparallel β-pleated sheet and bind the insecticide DDT was designed and synthesized. The synthetic peptide aggregated in 1 M acetic acid but was monomeric in aqueous 50% ethanol. In the latter solvent the 24-residue polypeptide and DDT formed a complex with an apparent dissociation constant of ≈2 × 10 −5 M. The DDT binding of an analogue of this peptide possessing the same amino acid residues in a random sequence was more than 2 orders of magnitude and that of bovine serum albumin at least 3 orders of magnitude weaker. The designed polypeptide could be crystallized.

关键词： protein secondary structure prediction Model building protein design Synthetic DDT-binding polypeptide Crystallization, of artificial polypeptide DDT 2,2-bis(p-chlorophenyl)-1,1,,1-trichloroethane

来源：评论

学校读者我要写书评

暂无评论

A NOVEL-APPROACH TO prediction OF THE 3-DIMENSIONAL structureS OF protein BACKBONES BY NEURAL NETWORKS

引用

FEBS LETTERS 1990年第1期261卷 43-46页

作者： BOHR, H BOHR, J BRUNAK, S COTTERILL, RMJ FREDHOLM, H LAUTRUP, B PETERSEN, SB NOVO IND AS NORDISK RES INSTNOVO ALLEDK-2880 BAGSVAERDDENMARK RISO NATL LAB DK-4000 ROSKILDEDENMARK TECH UNIV DENMARK DK-2800 LYNGBYDENMARK UNI C DK-2800 LYNGBYDENMARK NIELS BOHR INST DK-2100 COPENHAGEN 0DENMARK

Three-dimensional structures of protein backbones have been predicted using neural networks. A feed forward neural network was trained on a class of functionally, but not structurally, homologous proteins, using backpropagation learning. The network generated tertiary structure information in the form of binary distance constraints for the C(alpha) atoms in the protein backbone. The binary distance between two C(alpha) atoms was 0 if the distance between them was less than a certain threshold distance, and 1 otherwise. The distance constraints predicted by the trained neural network were utilized to generate a folded conformation of the protein backbone, using a steepest descent minimization approach.

关键词： protein secondary structure prediction protein tertiary structure prediction Neural network Binary distance matrix Steepest descent minimization

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：