Secondary Structure Assignment Research Articles

Suppressing sidechain modes and improving structural resolution for 2D IR spectroscopy via vibrational lifetimes.

Vibrational spectroscopy of protein structure often utilizes 13C18O-labeling of backbone carbonyls to further increase structural resolution. However, sidechains such as arginine, aspartate, and glutamate absorb within the same spectral region, complicating the analysis of isotope-labeled peaks. In this study, we report that the waiting time between pump and probe pulses in two-dimensional infrared spectroscopy can be used to suppress sidechain modes in favor of backbone amide I' modes based on differences in vibrational lifetimes. Furthermore, differences in the lifetimes of 13C18O-amide I' modes can aid in the assignment of secondary structure for labeled residues. Using model disordered and β-sheet peptides, it was determined that while β-sheets exhibit a longer lifetime than disordered structures, amide I' modes in both secondary structures exhibit longer lifetimes than sidechain modes. Overall, this work demonstrates that collecting 2D IR data at delayed waiting times, based on differences in vibrational lifetime between modes, can be used to effectively suppress interfering sidechain modes and further identify secondary structures.

Amino acid propensities for secondary structures and its variation across protein structures using exhaustive PDB data

Amino acid propensities for protein secondary structures are vital for protein structure prediction, understanding folding, and design, and have been studied using various theoretical and experimental methods. Traditional assessments of average propensities using statistical methods have been done on relatively smaller dataset for only a few secondary structures. They also involve averaging out the environmental factors and lack insights into consistency of preferences across diverse protein structures. While a few studies have explored variations in propensities across protein structural classes and folds, exploration of such variations across protein structures remains to be carried out. In this work, we have revised the average propensities for all six different secondary structures, namely α-helix, β-strand, 310-helix, π-helix, turn and coil, analyzing the most exhaustive dataset available till date using two robust secondary structure assignment algorithms, DSSP and STRIDE. The propensities evaluated here can serve as a standard reference. Moreover, we present here, for the first time, the propensities within individual protein structures and investigated how the preferences of residues and more interestingly, of their groups formed based on their structural features, vary across different unique structures. We devised a novel approach- the minimal set analysis, based on the propensity distribution of residues, which along with the group propensities led us to the conclusion that a residue's preference for a specific secondary structure is primarily dictated by its side chain's structural features. The findings in this study provide a more insightful picture of residues propensities and can be useful in protein folding and design studies.

Backbone chemical shift and secondary structure assignments for mouse siderocalin

The lipocalin protein family is a structurally conserved group of proteins with a variety of biological functions defined by their ability to bind small molecule ligands and interact with partner proteins. One member of this family is siderocalin, a protein found in mammals. Its role is discussed in inflammatory processes, iron trafficking, protection against bacterial infections and oxidative stress, cell migration, induction of apoptosis, and cancer. Though it seems to be involved in numerous essential pathways, the exact mechanisms are often not fully understood. The NMR backbone assignments for the human siderocalin and its rat ortholog have been published before. In this work we describe the backbone NMR assignments of siderocalin for another important model organism, the mouse - data that might become important for structure-based drug discovery. Secondary structure elements were predicted based on the assigned backbone chemical shifts using TALOS-N and CSI 3.0, revealing a high content of beta strands and one prominent alpha helical region. Our findings correlate well with the known crystal structure and the overall conserved fold of the lipocalin family.

Structural and functional analysis of engineered antibodies for cancer immunotherapy: insights into protein compactness and solvent accessibility

Antibodies are crucial tools in various biomedical applications, including immunotherapy. In this study, we focused on designing and engineering antibodies to enhance their structural dynamics and functional properties. By employing advanced computational techniques and experimental validation, we gained crucial insights into the impact of specific mutations on the engineered antibodies. This study investigates the design and engineering of antibodies to improve their structural dynamics and functional properties. Structural attributes, such as protein compactness and solvent accessibility, were assessed, revealing interesting trends in anti-CD3 and anti-HER2 antibodies. Mutations in CD3 antibodies resulted in a more stable conformation, while mutant HER2 antibodies exhibited altered interaction with the target. Analysis of secondary structure assignments demonstrated significant changes in the folding and stability of the mutant antibodies compared to the wild-type counterparts. The conformational landscape of the engineered antibodies was explored, providing insights into folding pathways and binding mechanisms. Overall, the current study highlights the significance of antibody design and engineering in modulating structural dynamics and functional properties. The findings contribute to developing improved immunotherapeutic strategies by optimising antibody-based therapeutics for targeted diseases with enhanced efficacy and precision. Communicated by Ramaswamy H. Sarma

Interactions and characterization of Zein-BSA nanoparticles: Multi-spectral analysis and molecular simulations

Zein-bovine serum albumin (BSA) nanoparticles (NPs) with varying mass ratios were prepared via anti-solvent precipitation method at pH 4.0. Zein-BSA NPs at high Zein mass ratios exhibited superior particle characteristics and surface morphology. Ultraviolet-visible, fluorescence, and X-ray diffraction analyses demonstrated the formation and conformational transition during the binding of Zein and BSA, while ANS hydrophobic binding, fourier transform infrared spectra, and circular dichroism confirmed the hydrophobic exposure and secondary structure assignment. Molecular dynamic simulations revealed that Zein spontaneously bound to the IIB domain of BSA with a binding free energy of −63.52 ± 6.37 kcal/mol, which was dominated by hydrophobic and electrostatic interactions, supplemented by hydrogen bonds. Dissociation experiments supported the key role of hydrophobic and electrostatic interactions in maintaining the stability of the NPs. These findings provided insights into the formation and stabilization mechanisms of Zein-BSA NPs, and their potential applications as encapsulation carriers for bioactive substances or as novel nanoscale food substrates.

Techniques for Developing Reliable Machine Learning Classifiers Applied to Understanding and Predicting Protein:Protein Interaction Hot Spots.

With machine learning now transforming the sciences, successful prediction of biological structure or activity is mainly limited by the extent and quality of data available for training, the astute choice of features for prediction, and thorough assessment of the robustness of prediction on a variety of new cases. In this chapter, we address these issues while developing and sharing protocols to build a robust dataset and rigorously compare several predictive classifiers using the open-source Python machine learning library, scikit-learn. We show how to evaluate whether enough data has been used for training and whether the classifier has been overfit to training data. The most telling experiment is 500-fold repartitioning of the training and test sets, followed by prediction, which gives a good indication of whether a classifier performs consistently well on different datasets. An intuitive method is used to quantify which features are most important for correct prediction.The resulting well-trained classifier, hotspotter, can robustly predict the small subset of amino acid residues on the surface of a protein that are energetically most important for binding a protein partner: the interaction hot spots. Hotspotter has been trained and tested here on a curated dataset assembled from 1046 non-redundant alanine scanning mutation sites with experimentally measured change in binding free energy values from 97 different protein complexes; this dataset is available to download. The accessible surface area of the wild-type residue at a given site and its degree of evolutionary conservation proved the most important features to identify hot spots. A variant classifier was trained and validated for proteins where only the amino acid sequence is available, augmented by secondary structure assignment. This version of hotspotter requiring fewer features is almost as robust as the structure-based classifier. Application to the ACE2 (angiotensin converting enzyme 2)receptor, which mediates COVID-19 virus entry into human cells, identified the critical hot spot triad of ACE2 residues at the center of the small interface with the CoV-2 spike protein. Hotspotter results can be used to guide the strategic design of protein interfaces and ligands and also to identify likely interfacial residues for protein:protein docking.

DoubleHelix: nucleic acid sequence identification, assignment and validation tool for cryo-EM and crystal structure models.

Sequence assignment is a key step of the model building process in both cryogenic electron microscopy (cryo-EM) and macromolecular crystallography (MX). If the assignment fails, it can result in difficult to identify errors affecting the interpretation of a model. There are many model validation strategies that help experimentalists in this step of protein model building, but they are virtually non-existent for nucleic acids. Here,I present doubleHelix-a comprehensive method for assignment, identification, and validation of nucleic acid sequences in structures determined using cryo-EM and MX. The method combines a neural network classifier of nucleobase identities and a sequence-independent secondary structure assignment approach. I show that the presented method can successfully assist sequence-assignment step in nucleic-acid model building at lower resolutions, where visual map interpretation is very difficult. Moreover, I present examples of sequence assignment errors detected using doubleHelix in cryo-EM and MX structures of ribosomes deposited in the Protein Data Bank, which escaped the scrutiny of available model-validation approaches. The doubleHelix program source code is available under BSD-3 license at https://gitlab.com/gchojnowski/doublehelix.

De novo protein fold design through sequence-independent fragment assembly simulations

De novo protein design generally consists of two steps, including structure and sequence design. Many protein design studies have focused on sequence design with scaffolds adapted from native structures in the PDB, which renders novel areas of protein structure and function space unexplored. We developed FoldDesign to create novel protein folds from specific secondary structure (SS) assignments through sequence-independent replica-exchange Monte Carlo (REMC) simulations. The method was tested on 354 non-redundant topologies, where FoldDesign consistently created stable structural folds, while recapitulating on average 87.7% of the SS elements. Meanwhile, the FoldDesign scaffolds had well-formed structures with buried residues and solvent-exposed areas closely matching their native counterparts. Despite the high fidelity to the input SS restraints and local structural characteristics of native proteins, a large portion of the designed scaffolds possessed global folds completely different from natural proteins in the PDB, highlighting the ability of FoldDesign to explore novel areas of protein fold space. Detailed data analyses revealed that the major contributions to the successful structure design lay in the optimal energy force field, which contains a balanced set of SS packing terms, and REMC simulations, which were coupled with multiple auxiliary movements to efficiently search the conformational space. Additionally, the ability to recognize and assemble uncommon super-SS geometries, rather than the unique arrangement of common SS motifs, was the key to generating novel folds. These results demonstrate a strong potential to explore both structural and functional spaces through computational design simulations that natural proteins have not reached through evolution.

Protein secondary structure assignment using residual networks.

Proteins are constructed from amino acid sequences. Their structural classifications include primary, secondary, tertiary, and quaternary, with tertiary and quaternary structures influencing protein function. Because a protein's structure is inextricably connected to its biological function, machine learning algorithms that can better anticipate the structures have the potential to lead to new scientific discoveries in human health and improve our capacity to develop new treatments. Protein secondary structure assignment enriches the structural and functional understanding of proteins. It helps in proteinstructure comparison and classification studies, besides facilitating secondary and tertiary structure prediction systems. Several secondary structure assignment methods have been developed since the 1980s, most of which are based on hydrogen bond analysis and atomic coordinate features. However, the assignment process becomes complex when protein data includes missing atoms. Deep neural networks are often referred to as universal function approximators because they can approximate any function to produce the desired output when properly designed and trained. Optimised deep learning architectures have already proven their ability to increase performance in a wide range of problems. Recently, the ResNet architecture has garnered significant interest due to its applicability in various areas, including image classification and protein contact map prediction. The proposed model, which is based on the ResNet architecture, assigns secondary structures using Cα atom coordinates. The model achieved an accuracy of 94% when evaluated against the benchmark and independent test sets. The findings encourage the development of new deep learning-based methods that are more generalised across various protein learning tasks. Furthermore, it allows computational biologists to delve deeper into integrating these techniques with experimental methods. The model codes are available at: https://github.com/jisnava/ResNet_for_Structure_Assignments/ .

Automated Protein Secondary Structure Assignment from Cα Positions Using Neural Networks.

The assignment of secondary structure elements in protein conformations is necessary to interpret a protein model that has been established by computational methods. The process essentially involves labeling the amino acid residues with H (Helix), E (Strand), or C (Coil, also known as Loop). When particular atoms are absent from an input protein structure, the procedure becomes more complicated, especially when only the alpha carbon locations are known. Various techniques have been tested and applied to this problem during the last forty years. The application of machine learning techniques is the most recent trend. This contribution presents the HECA classifier, which uses neural networks to assign protein secondary structure types. The technique exclusively employs C coordinates. The Keras (TensorFlow) library was used to implement and train the neural network model. The BioShell toolkit was used to calculate the neural network input features from raw coordinates. The study’s findings show that neural network-based methods may be successfully used to take on structure assignment challenges when only C trace is available. Thanks to the careful selection of input features, our approach’s accuracy (above 97%) exceeded that of the existing methods.

CSSR: assignment of secondary structure to coarse-grained RNA tertiary structures.

RNA secondary-structure (rSS) assignment is one of the most routine forms of analysis of RNA 3D structures. However, traditional rSS assignment programs require full-atomic structures of the individual RNA nucleotides. This prevents their application to the modeling of RNA structures in which base atoms are missing. To address this issue, Coarse-grained Secondary Structure of RNA (CSSR), an algorithm for the assignment of rSS for structures in which nucleobase atomic positions are incomplete, has been developed. Using CSSR, an rSS assignment accuracy of ∼90% is achieved even for RNA structures in which only one backbone atom per nucleotide is known. Thus, CSSR will be useful for the analysis of experimentally determined and computationally predicted RNA 3D structures alike. The source code of CSSR is available at https://github.com/pylelab/CSSR.

An End-to-End Deep Learning Pipeline for Assigning Secondary Structure in Proteins

Protein secondary structure assignment, a subdiscipline of computational chemistry, is yet to be explored using deep learning techniques. Protein secondary structure elements are assigned to support structural analysis and prediction. Algorithms like DSSP, generally regarded as the gold standard for assigning the secondary structure of proteins, need full atom information to label protein coordinates. The PDB database has been the major repository for data on the 3D structures of proteins, nucleic acids, and other complex assemblies since 1971. However, a significant fraction of protein structures contains missing atoms. As a result, new approaches to reliably and consistently assigning secondary structures based on coarse-grained atomic coordinates are needed. While deep learning architectures have an unparalleled track record in applications such as protein structure prediction, there are only a few known deep learning solutions for structure assignment problems. While the gold standard methods are based on bonding information and other geometric characteristics, deep learning methods extract features themselves without human intervention. The benefit that standardised datasets provide to the effectiveness of deep learning systems in multiple domains motivated us to create a labeled dataset for protein structure assignment tasks. Additionally, a deep learning model, named PSSADL, solely based on [Formula: see text] coordinates was trained on the generated dataset to validate its potency. The proposed method, which combines Convolutional Neural Networks (CNN) and Long Short Term Memory (LSTM)/Bidirectional LSTM networks, has been compared to the established standards and more recent techniques. The model achieved an accuracy of [Formula: see text] on the benchmark and individual test sets. The results show that deep learning techniques have a promising future in protein structure analysis, implying that the dataset developed as part of our work will be a valuable resource for further protein structure research.

Molecular Modeling Strategies of Cancer Multidrug Resistance.

Cancer remains a leading cause of morbidity and mortality worldwide. Hence, the increase in cancer cases observed in the elderly population, as well as in children and adolescents, makes human malignancies a prime target for anticancer drug development. Although highly effective chemotherapeutic agents are continuously developed and approved for clinical treatment, the major impediment towards curative cancer therapy remains multidrug resistance (MDR). In recent years, intensive studies have been carried out on the identification of new therapeutic molecules to reverse MDR efflux transporters of the ATP-binding cassette (ABC) superfamily. Although a great deal of progress has been made in the development of specific inhibitors for certain MDR efflux pumps in experimental studies, advanced computational studies can accelerate this drug development process. In the literature, there are many experimental studies on the impact of natural products and synthetic small molecules on the reversal of cancer MDR. Molecular modeling methods provide an opportunity to explain the activity of these molecules on the ABC-transporter family with non-covalent interactions as well as it is possible to carry out studies for the discovery of new anticancer drugs specific to MDR with these methods. The coordinate file of the 3-dimensional (3D) structure of the target protein is indispensable for molecular modeling studies. In some cases where a 3D structure cannot be obtained by experimental methods, the homology modeling method can be applied to obtain the file containing the target protein's information including atomic coordinates, secondary structure assignments, and atomic connectivity. Homology modeling studies are of great importance for efflux transporter proteins that still lack 3D structures due to crystallization problems with multiple hydrophobic transmembrane domains. Quantum mechanics, molecular docking and molecular dynamics simulation applications are the most frequently used molecular modeling methods in the literature to investigate non-covalent interactions between the drug-ABC transporter superfamily. The quantitative structure-activity relationship (QSAR) model provides a relationship between the chemical properties of a compound and its biological activity. Determining the pharmacophore region for a new drug molecule by superpositioning a series of molecules according to their physicochemical properties using QSAR models is another method in which molecular modeling is used in computational drug development studies with ABC transporter proteins. There are also in silico absorption, distribution, metabolism, excretion, and toxicity (ADME/Tox) studies conducted to make a prediction about the pharmacokinetic properties, and drug-likeness of new molecules. Drug repurposing studies, which have become a trending topic in recent years, involve identifying possible new targets for an already approved drug molecule. There are few studies in the literature in which drug repurposing performed by molecular modelling methods has been applied on ABC transporter proteins. The aim of the current paper is to create a complete review of drug development studies including aforementioned molecular modeling methods carried out between the years 2019-2021. Furthermore, an intensive investigation is also conducted on licensed applications and free web servers used in in silico studies. The current review is an up-to-date guide for researchers who plan to conduct computational studies with MDR transporter proteins.

Assigning secondary structure in proteins using AI.

Knowledge about protein structure assignment enriches the structural and functional understanding of proteins. Accurate and reliable structure assignment data is crucial for secondary structure prediction systems. Since the 1980s, various methods based on hydrogen bond analysis and atomic coordinate geometry, followed by machine learning, have been employed in protein structure assignment. However, the assignment process becomes challenging when missing atoms are present in the protein files. Our method proposed a multi-class classifier program named DLFSA for assigning protein secondary structure elements (SSE) using convolutional neural networks (CNNs). A fast and efficient GPU-based parallel procedure extracts fragments from protein files. The model implemented in this work is trained with a subset of the protein fragments and achieves 88.1% and 82.5% train and test accuracy, respectively. The model uses only Cα coordinates for secondary structure assignments. The model has been successfully tested on a few full-length proteins also. Results from the fragment-based studies demonstrate the feasibility of applying deep learning solutions for structure assignment problems.

A Comprehensive Study of Human Serum Albumin Interaction With Trimethoprim Using Molecular Docking and Molecular Dynamics Methods: An Appropriate Tool for Drug Delivery Systems

Background: Human Serum Albumin (HSA) is one of the most prominent proteins in human blood. Trimethoprim (TMP) is an efficient antibiotic drug for treating pneumocystis pneumonia. Patients with HIV/AIDS and cancer are highly affected by this disease due to immune system deficiency. Objective: This study aims to evaluate the Molecular Dynamics (MD) simulation of HSA with TMP for drug delivery systems. Methods: In the first step, the 3D structure of HSA and TMP were determined by PDB (Protein Data Bank) and PubChem, respectively. Then, the molecular docking was done via AutoDock Vina software, and the best complex was selected based on the lowest binding energy. Finally, the structural characteristics of the above complex were evaluated. Results: The results showed that TMP binds to the HSA molecule with a binding energy of -7.3 kcal/mol, and this binding causes changes in the third and second structures of the HSA. Thus, Root-Mean-Square Deviation (RMSD) and radius of gyration results proved the third structural change, and the results obtained from DSSP (Database of Secondary Structure assignment for all Protein entries) confirmed the second structural modification. The TMP-HSA complex formation is accompanied by hydrophobic interaction between residues of Tyr150, Ala291, His288, Leu238, Leu219, Lys199, Lys195, Glu153, and TMP. The TMP molecule had two hydrogen bonds with Arg222 residue and three with Ser192. Furthermore, the final PDB file of the MD simulation process showed that the TMP molecule reacted with HSA (IIA chain). Conclusion: Because of the extensive application of TMP in infectious diseases and appropriate interaction with HSA, the complex could be used for the purposeful transport of nanoparticles in the future.

Secondary structure assignment of proteins in the absence of sequence information.

The structure of proteins is organized in a hierarchy among which the secondary structure elements, α-helix, β-strand and loop, are the basic bricks. The determination of secondary structure elements usually requires the knowledge of the whole structure. Nevertheless, in numerous experimental circumstances, the protein structure is partially known. The detection of secondary structures from these partial structures is hampered by the lack of information about connecting residues along the primary sequence. We introduce a new methodology to estimate the secondary structure elements from the values of local distances and angles between the protein atoms. Our method uses a message passing neural network, named Sequoia, which allows the automatic prediction of secondary structure elements from the values of local distances and angles between the protein atoms. This neural network takes as input the topology of the given protein graph, where the vertices are protein residues, and the edges are weighted by values of distances and pseudo-dihedral angles generalizing the backbone angles and ψ. Any pair of residues, independently of its covalent bonds along the primary sequence of the protein, is tagged with this distance and angle information. Sequoia permits the automatic detection of the secondary structure elements, with an F1-score larger than 80% for most of the cases, when α helices and β strands are predicted. In contrast to the approaches classically used in structural biology, such as DSSP, Sequoia is able to capture the variations of geometry at the interface of adjacent secondary structure element. Due to its general modeling frame, Sequoia is able to handle graphs containing only atoms, which is particularly useful on low resolution structural input and in the frame of electron microscopy development. Sequoia source code can be found at https://github.com/Khalife/Sequoia with additional documentation. Supplementary data are available at Bioinformatics Advances online.

Protein secondary structure assignment using pc-polyline and convolutional neural network.

The assignment of protein secondary structure elements (SSEs) underpins structural analysis and prediction. The backbone of a protein could be adequately represented using a pc-polyline that passes through the centers of its peptide planes. One salient feature of pc-polyline representation is that the secondary structure of a protein becomes recognizable in a matrix whose elements are the pairwise distances between two peptide plane centers. Thus, a pc-polyline could in turn be used to assign SSEs. Using convolutional neural network (CNN) here we confirm that a pc-polyline indeed contains enough information for it to be used for the accurate assignments of the six SSE types: α-helix, β-sheet, β-bulge, 310 -helix, turn and loop. The applications to three large data sets show that the assignments by our CNN-based p2psse program agree very well with those by dssp, stride and quite well with those by five other programs. The analyses of their SSE assignments raise some general questions about the characterizations of protein secondary structure. In particular the analyses illustrate the difficulty with giving a quantitative and consistent definition for each of the six SSE types especially for 310 -helix, β-bulge, turn or loop in terms of either backbone H-bond patterns, or backbone dihedral angles, or Cα -polyline or pc-polyline. The difficulty suggests that the SSE space though being dominated by the regions for the six SSE types is to a certain degree continuous. The program is available at https://github.com/wlincong/p2pSSE.

Detecting protein and DNA/RNA structures in cryo-EM maps of intermediate resolution using deep learning

An increasing number of density maps of macromolecular structures, including proteins and DNA/RNA complexes, have been determined by cryo-electron microscopy (cryo-EM). Although lately maps at a near-atomic resolution are routinely reported, there are still substantial fractions of maps determined at intermediate or low resolutions, where extracting structure information is not trivial. Here, we report a new computational method, Emap2sec+, which identifies DNA or RNA as well as the secondary structures of proteins in cryo-EM maps of 5 to 10 Å resolution. Emap2sec+ employs the deep Residual convolutional neural network. Emap2sec+ assigns structural labels with associated probabilities at each voxel in a cryo-EM map, which will help structure modeling in an EM map. Emap2sec+ showed stable and high assignment accuracy for nucleotides in low resolution maps and improved performance for protein secondary structure assignments than its earlier version when tested on simulated and experimental maps.

PSIQUE: Protein Secondary Structure Identification on the Basis of Quaternions and Electronic Structure Calculations

The secondary structure is important in protein structure analysis, classification, and modeling. We have developed a novel method for secondary structure assignment, termed PSIQUE, based on the potential energy surface (PES) of polyalanine obtained using an infinitely long chain model and density functional theory calculations. First, uniform protein segments are determined in terms of a difference of quaternions between neighboring amino acids along the protein backbone. Then, the identification of the secondary structure motifs is carried out based on the minima found in the PES. PSIQUE shows good agreement with other secondary structure assignment methods. However, it provides better discrimination of subtle secondary structures (e.g., helix types) and termini and produces more uniform segments while also accounting for local distortions. Overall, PSIQUE provides a precise and reliable assignment of secondary structures, so it should be helpful for the detailed characterization of the protein structure.

Impact of protein dynamics on secondary structure prediction

Protein 3D structures support their biological functions. As the number of protein structures is negligible in regards to the number of available protein sequences, prediction methodologies relying only on protein sequences are essential tools. In this field, protein secondary structure prediction (PSSPs) is a mature area, and is considered to have reached a plateau.Nonetheless, proteins are highly dynamical macromolecules, a property that could impact the PSSP methods. Indeed, in a previous study, the stability of local protein conformations was evaluated demonstrating that some regions easily changed to another type of secondary structure.The protein sequences of this dataset were used by PSSPs and their results compared to molecular dynamics to investigate their potential impact on the quality of the secondary structure prediction. Interestingly, a direct link is observed between the quality of the prediction and the stability of the assignment to the secondary structure state. The more stable a local protein conformation is, the better the prediction will be. The secondary structure assignment not taken from the crystallized structures but from the conformations observed during the dynamics slightly increase the quality of the secondary structure prediction. These results show that evaluation of PSSPs can be done differently, but also that the notion of dynamics can be included in development of PSSPs and other approaches such as de novo approaches.