Combining PSSM and physicochemical feature for protein structure prediction with support vector machine

Protein structure prediction is one of the most important problems in modern computational biology. Protein secondary structure prediction is a key step in prediction of protein tertiary structure. There have emerged many methods based on machine learning techniques, such as neural networks (NN) and support vector machine (SVM) etc., to focus on the prediction of the secondary structures. In this paper, a new method was proposed based on SVM. Different from the existing methods, this method takes into account of the physical-chemical properties and structure properties of amino acids. When tested on the most popular dataset CB513, it achieved a Q(3) accuracy of 0.7844, which illustrates that it is one of the top range methods for protein of secondary structure prediction.

φ, ψ Representations of three-dimensional structures of 125 globular proteins were depicted for analyzing conformational details in their secondary and tertiary structures. They can be drawn in the form of a two-dimensional diagram. The new type of representation of the protein tertiary structure can be used as a stereotyped finger printing for 125 protein molecules. It is powerful for a precise comparison of the relationship of the primary, secondary, and tertiary structures of homologous proteins. The precise or broad similarity between the patterns of proteins classified into the same family is obvious. This representation is also useful in readily recognizing all of the secondary structures as one progresses along the chain. Complicated three-dimensional structures of 125 globular proteins are easily recognized to be built up from only four secondary structures: helix, β-strand, β-turn, and unordered structure. A combination of intersegments hydrogen bonds among β-strands was also shown in the diagram. In order to demonstrate the usefulness of the diagram, 195 type-II β-turns were extracted from 125 proteins using the new type of two-dimensional φ, ψ diagrams. Their amino acid sequences were surveyed so as better to understand amino acid preferences previously observed. For their analysis, a normalized preference (NP) value of an amino acid residue at a particular position of the secondary structure is defined as the ratio of the average percentage composition at the position based on average percentage composition at large. The statistical significance of the NP-value used in this study is more simple and clearer than that of the d-test based on a normal distribution, which was used in the previous study. It is clearly shown that a high NP-value (5.0) of a single P residue for the position i + 1 of the type-II β-turn is an average preference of a variety of P residues in the middle of triplets consisting of consecutive amino acid residues. The NP-values of P residues, in the middle of 25 particular triplets are outstandingly high. Out of 400 possible P residues 322 kinds of P residues were found in 125 proteins. A remarkable high NP-value (8.8) of a single G residue for the position i + 2 is also an average preference of G residues in the middle of various triplets.

2 - The Physiological Structure and Function of Proteins

Chapter 1 of this thesis provides an overview of the structure and function of proteins as well as their investigation via NMR spectroscopy. Proteins are one of the most important building blocks of life. They are part of every organism and fulfill a variety of functions including, for example, the catalysis of many chemical reactions, which is the reason various processes in an organism become possible in the first place. Proteins are made up of amino acids, which are linked together via peptide bonds during the protein biosynthesis. The 22 proteinogenic amino acids result in an almost unlimited number of sequential combinations that give proteins their versatility. The sequence of amino acids in a protein is also called the primary structure. From this, locally limited structural motifs such as α-helices and β-sheets are formed, representing the secondary structure. Further intramolecular interactions create the tertiary structure of a protein, which can be seen as the overall three-dimensional structure of the protein. Furthermore, several individual proteins can assemble into a larger complex to form a quaternary structure. There are various methods for investigating the structure of proteins, including nuclear magnetic resonance (NMR) spectroscopy. In NMR spectroscopy, atomic nuclei with a non-zero nuclear spin are polarized by a constant magnetic field and then excited using radio frequency pulses. The emitted resonance frequencies are detected in the form of a free induction decay (FID) and converted into an NMR spectrum by a Fourier transformation. Many NMR experiments are available for researching proteins. By default, a two-dimensional 1H15N-HSQC spectrum is measured, which provides signals for each N-H bond in a protein meaning that one signal is obtained for each N-H in the backbone of the amino acids. Using 3D NMR experiments, these signals can then be assigned to the respective amino acids in the protein sequence. This makes it possible to investigate the dynamics of proteins as well as the formation of complexes with a ligand or another biomacromolecule if they interact. Further 3D NMR experiments also make it possible to assign the side-chains of the protein. With the help of the determined resonance assignments and spatial information from various NOESY spectra, a protein structure can then be calculated using a suitable computer program. Recently, the prediction of protein structures using AI (artificial intelligence)-supported programs has also become more prevalent. In particular, the artificial neural network AlphaFold has been extremely successful in predicting correct structures without any experimental data provided by the user. Chapter 2 of this thesis introduces the research field of small proteins. These are proteins with a length of usually less than 100 amino acids, although definitions in literature vary and upper limits of, for example, 50 or 200 amino acids exist as well. Various other names are also used to describe small proteins, such as μ-proteins or sORF-encoded peptides. The field of small proteins is still relatively new, as they were overlooked in research for many years. One reason for this was the lower limit of 100 codons that was set for the identification of open reading frames (ORFs) in genome projects. This served to exclude numerous biologically irrelevant ORFs from genome annotation. However, this meant that functional ORFs coding for small proteins were also ignored. Another reason was that experimental methods were not designed for the analysis of small proteins, meaning they could often not be detected at all. However, new analytical methods such as ribosome profiling and a generally increased focus on the detection of small proteins have led to their detection in all domains of life. In the research group of Prof. Dr. Schwalbe, small proteins of various prokaryotes were structurally characterized as part of a special priority program by the DFG. NMR spectroscopy was primarily used for this purpose, as it is the method of choice for the investigation of small proteins. The results of this research were published in several articles, two of which are part of this dissertation.

Background and Objectives: Changes in the secondary and tertiary structures of protein molecules during thermal denaturation lead to changes in their vibrational spectra. Vibrations corresponding to elements of the primary and secondary structure of the protein are manifested in the fingerprint range, while vibrational modes of large fragments corresponding to higher levels of the protein structural hierarchy can be observed in the low-frequency (40–500 cm−1) spectral range. The purpose of this work is to reveal changes in the vibrational spectra of proteins resulting from denaturation that can be used to characterize changes in the protein structure. Materials and Methods: Two proteins (collagen and chymotrypsin) having different secondary and tertiary structures are studied using Raman spectroscopy. The experimental data are analyzed using bandpass filtering of the Fourier transforms of the spectral curves. Results: Vibrational spectra of two proteins (collagen and chymotrypsin) having different secondary and tertiary structures, as well as the spectra of thermally denatured samples of these proteins have been measured in the fingerprint and low-frequency ranges. Several low-frequency spectral features that can be used to characterize structural changes of protein molecules have been considered. A few bands may correspond to vibrations of tertiary structure elements (both in the low-frequency range and in the “fingerprint” range). Conclusion: A comparison of the vibrational spectra of native and denatured (superhelical) collagen, as well as native and denatured (globular) chymotrypsin shows that changes in the amide I and amide III bands sensitive to the secondary structure are supplemented with spectral changes in bands that are not assigned to elements of the secondary structure and can be related to changes at higher levels of structural hierarchy. Comparison of the low-frequency vibrational spectra indicates a lower sensitivity of presumably tertiary structure of the globular protein to denaturation.

Protein adsorption at oil-water interfaces interfacial structure and function of proteins in emulsions

In this study Haar features are extracted from images of sub-sequences of amino acid and classified by Support Vector Machine (SVM). We apply a novel approach of integration Haar-like features extraction from primary protein structure to predication three states of secondary protein structure. The sequences of primary protein are divided into different slides windows for representation images and then Haar-like feature have been extracted from these images to classify three-category of secondary protein structure helix (H), strand (E) and coil (C). The final prediction results were generated from SVM overall per residue accuracies are: - accuracy of helix(H) reached 83.93%, accuracy of Sheet(E) is 85.15% and accuracy of the Coil (C) is 81.0126 %. Images are scanned from amino acid sequences are specified by the selection window sizes, when the size of window is small the important information of predicting secondary structure relay outside the window. It has taken only local sequence information. When the size of window has been increased the performance has been deteriorated. Haar-like gives a perfect input data of SVM with a huge amount of data, also for improvement of support vector machine are used varying cost parameters.

Prediction of protein structure classes by incorporating different protein descriptors into general Chou’s pseudo amino acid composition

The secondary structure of proteins is significant for studying the three-dimensional structure and functions of proteins. Several models from image understanding and natural language modeling have been successfully adapted in the protein sequence study area, such as Long Short-term Memory (LSTM) network and Convolutional Neural Network (CNN). Recently, Gated Convolutional Neural Network (GCNN) has been proposed for natural language processing and reduces latency while achieving high levels of sentence scoring. Conditionally Parameterized Convolution (CondConv), which use extra sample-dependant modules to conditionally adjust the convolutional network, have achieved great success in the image processing area. In this paper, we propose a novel Conditionally Parameterized Convolutional network (CondGCNN) which utilize the power of both CondConv and GCNN, and we leverage an ensemble encoder to combine the capabilities of both LSTM and CondGCNN to encode protein sequences to obtain better sequential features from proteins. In addition, due to the similarity between the image segmentation problem and the secondary structure prediction problem, we propose an ASP network (Atrous Spatial Pyramid Pooling (ASPP) based network) as the secondary structure generator in our proposed framework. We have conducted extensive ablation studies over each component in the proposed model to verify its effectiveness. Extensive experiments show that the proposed method can achieve higher performance on protein secondary structure prediction task than existing methods on CB513, Caspll and CASP12 datasets. Our method is expected to be useful for protein structure and further protein functions prediction.

Independent isolates of human immunodeficiency virus (HIV) exhibit a striking genomic diversity, most of which is located in the viral envelope gene. Since this property of the HIV group of viruses may play an important role in the pathobiology of the virus, we analyzed the predicted amino acid sequences of the envelope proteins of seven different HIV strains, three of which represent sequential isolates from a single patient. By using a computer program that predicts the secondary protein structure and superimposes values for hydrophilicity, surface probability, and flexibility, we identified several potential antigenic epitopes in the envelope proteins of the seven different viruses. Interestingly, the majority of the predicted epitopes in the exterior envelope protein (gp120) were found in regions of high sequence variability which are interspersed with highly conserved regions among the independent viral isolates. A comparison of the sequential viral isolates revealed that changes concerning the secondary structure of the protein occurred only in regions which were predicted to be antigenic, predominantly in highly variable regions. The membrane-associated protein gp41 contains no highly variable regions; about 80% of the amino acids were found to be conserved, and only one hydrophilic area was identified as likely to be accessible to antibody recognition. These findings give insight into the secondary and possible tertiary structure of variant HIV envelope proteins and should facilitate experimental approaches directed toward the identification and fine mapping of HIV envelope proteins.

ProteinShop Gives Researchers a Hands-On Tool for Manipulating, Visualizing Protein Structures. The Human Genome Project and other biological research efforts are creating an avalanche of new data about the chemical makeup and genetic codes of living organisms. But in order to make sense of this raw data, researchers need software tools which let them explore and model data in a more intuitive fashion. With this in mind, researchers at Lawrence Berkeley National Laboratory and the University of California, Davis, have developed ProteinShop, a visualization and modeling program which allows researchers to manipulate protein structures with pinpoint control, guided in large part by their own biological and experimental instincts. Biologists have spent the last half century trying to unravel the ''protein folding problem,'' which refers to the way chains of amino acids physically fold themselves into three-dimensional proteins. This final shape, which resembles a crumpled ribbon or piece of origami, is what determines how the protein functions and translates genetic information. Understanding and modeling this geometrically complex formation is no easy matter. ProteinShop takes a given sequence of amino acids and uses visualization guides to help generate predictions about the secondary structures, identifying alpha helices and flat beta strands, and the coil regions that bind them. Once secondary structures are in place, researchers can twist and turn these pre-configurations until they come up with a number of possible tertiary structure conformations. In turn, these are fed into a computationally intensive optimization procedure that tries to find the final, three-dimensional protein structure. Most importantly, ProteinShop allows users to add human knowledge and intuition to the protein structure prediction process, thus bypassing bad configurations that would otherwise be fruitless for optimization. This saves compute cycles and accelerates the entire process, so that more and larger problems can be attempted. Currently, the program designers are working to make ProteinShop more applicable and adaptable to different protein folding methodologies. If users could manipulate structures from a biological point of view, and then put them back in the queue for more optimization, the process of experimentation and discovery in protein research could be greatly enhanced. The group is also investigating the use of stereoscopic rendering and three-dimensional input devices to remove the limitations of a two-dimensional interface. Clearly, protein-folding research will have far-reaching ramifications. It could lead to new insights about diseases ranging from Alzheimer's to Cystic fibrosis, which scientists believe are caused by protein folding gone wrong. A better understanding of protein structures could also lead to the engineering of altogether new proteins, and shed light on how drugs bind proteins to alter their structure and function. Above all, ProteinShop is an important tool that will help scientists unravel one of the most challenging problems that theoretical and computational chemistry has to offer.

Experimental techniques capable of determining the structure and dynamics of proteins are continuously being developed in order to understand protein function. Among existing methods, site-directed spin labeling in combination with saturation recovery (SR) electron paramagnetic resonance spectroscopy contributes uniquely to the determination of secondary and tertiary protein structure under physiological conditions, independent of molecular weight and complexity. In addition, SR of spin labeled proteins was recently demonstrated to be sensitive to conformational exchange events with characteristic lifetimes on the order of μs, a time domain that presents a significant challenge to other spectroscopic techniques. In this chapter, we present the theoretical background necessary to understand the capabilities of SR as applied to spin labeled proteins, the instrumental requirements, and practical experimental considerations necessary to obtain interpretable data, and the use of SR to obtain information on protein: (1) secondary structure via solvent accessibility measurements, (2) tertiary structure using interspin distance measurements, and (3) conformational exchange.

Effects of different ultrasound powers on the structure and stability of protein from sea cucumber gonad

Unravelling the complex correlation between chemical shifts of (13) C (α),(13) C (β),(13) C',(1) H (α),(15) N,(1) H ( N ) atoms in amino acids of proteins from NMR experiment and local structural environments of amino acids facilitates the assignment of secondary structures of proteins. This is an important impetus for both determining the three-dimensional structure and understanding the biological function of proteins. The previous empirical correlation scores which relate chemical shifts of (13) C (α),(13) C (β),(13) C',(1) H (α),(15) N,(1) H ( N ) atoms to secondary structures resulted in progresses toward assigning secondary structures of proteins. However, the physical-mathematical framework for these was elusive partly due to both the limited and orthogonal exploration of higher-dimensional chemical shifts of hetero-nucleus and the lack of physical-mathematical understanding underlying those correlation scores. Here we present a simple multi-dimensional hetero-nuclear chemical shift score function (MDHN-CSSF) which captures systematically the salient feature of such complex correlations without any references to a random coil state of proteins. We uncover the symmetry-breaking vector and its reliability order not only for distinguishing different secondary structures of proteins but also for capturing the delicate sensitivity interplayed among chemical shifts of (13) C (α),(13) C (β),(13) C',(1) H (α),(15) N,(1) H ( N ) atoms simultaneously, which then provides a straightforward framework toward assigning secondary structures of proteins. MDHN-CSSF could correctly assign secondary structures of training (validating) proteins with the favourable (comparable) Q3 scores in comparison with those from the previous correlation scores. MDHN-CSSF provides a simple and robust strategy for the systematic assignment of secondary structures of proteins and would facilitate the de novo determination of three-dimensional structures of proteins.

Accurate prediction of protein structure relies heavily on exploiting multiple sequence alignment (MSA) for residue mutations and correlations as this information specifies protein tertiary structure. The widely used prediction approaches usually transform MSA into inter-mediate models, say position-specific scoring matrix or profile hidden Markov model. These inter-mediate models, however, cannot fully represent residue mutations and correlations carried by MSA; hence, an effective way to directly exploit MSAs is highly desirable. Here, we report a novel sequence set network (called Seq-SetNet) to directly and effectively exploit MSA for protein structure prediction. Seq-SetNet uses an 'encoding and aggregation' strategy that consists of two key elements: (i) an encoding module that takes a component homologue in MSA as input, and encodes residue mutations and correlations into context-specific features for each residue; and (ii) an aggregation module to aggregate the features extracted from all component homologues, which are further transformed into structural properties for residues of the query protein. As Seq-SetNet encodes each homologue protein individually, it could consider both insertions and deletions, as well as long-distance correlations among residues, thus representing more information than the inter-mediate models. Moreover, the encoding module automatically learns effective features and thus avoids manual feature engineering. Using symmetric aggregation functions, Seq-SetNet processes the homologue proteins as a sequence set, making its prediction results invariable to the order of these proteins. On popular benchmark sets, we demonstrated the successful application of Seq-SetNet to predict secondary structure and torsion angles of residues with improved accuracy and efficiency. The code and datasets are available through https://github.com/fusong-ju/Seq-SetNet. Supplementary data are available at Bioinformatics online.