Structure Prediction Modeling Research Articles

Molecular Modeling and Bioinformatics Analysis of Drug-Receptor Interactions in the System Formed by Glargine, Its Metabolite M1, the Insulin Receptor, and the IGF1 Receptor

Introduction:Insulin and insulin-like growth factor type 1 (IGF1) regulate multiple physiological functions by acting on the insulin receptor (IR) and insulin-like growth factor type 1 receptor (IGF1R). The insulin analog glargine differs from insulin in three residues (GlyA21, ArgB31, ArgB32), and it is converted to metabolite M1 (lacks residues ArgB31 and ArgB32) by in vivo processing. It is known that activation of these receptors modulates pathways related to metabolism, cell division, and growth. Though, the structures and structural basis of the glargine interaction with these receptors are not known.Aim:To generate predictive structural models, and to analyze the drug/receptor interactions in the system formed by glargine, its metabolite M1, IR, and IGF1R by using bioinformatics tools.Methods:Ligand/receptor models were built by homology modeling using SWISSMODEL, and surface interactions were analyzed using Discovery Studio® Visualizer. Target and hetero target sequences and appropriate template structures were used for modeling.Results:Our glargine/IR and metabolite M1/IR models showed an overall symmetric T-shaped conformation and full occupancy with four ligand molecules. The glargine/IR model revealed that the glargine residues ArgB31 and ArgB32 fit in a hydrophilic region formed by the α-chain C-terminal helix (αCT) and the cysteine-rich region (CR) domain of this receptor, close to the CR residues Arg270-Arg271-Gln272 and αCT residue Arg717. Regarding IGF1R, homologous ligand/receptor models were further built assuming that the receptor is in a symmetrical T-shaped conformation and is fully occupied with four ligand molecules, similar to what we described for IR. Our glargine/IGF1R model showed the interaction of the glargine residues ArgB31 and ArgB32 with Glu264 and Glu305 in the CR domain of IGF1R.Conclusion:Using bioinformatics tools and predictive modeling, our study provides a better understanding of the glargine/receptor interactions.

Secondary structure prediction of protein based on multi scale convolutional attention neural networks.

To fully extract the local and long-range information of amino acid sequences and enhance the effective information, this research proposes a secondary structure prediction model of protein based on a multi-scale convolutional attentional neural network. The model uses a multi-channel multi-scale parallel architecture to extract amino acid structure features of different granularity according to the window size. The reconstructed feature maps are obtained via multiple convolutional attention blocks. Then, the reconstructed feature map is fused with the input feature map to obtain the enhanced feature map. Finally, the enhanced feature map is fed to the Softmax classifier for prediction. While the traditional cross-entropy loss cannot effectively solve the problem of non-equilibrium training samples, a modified correlated cross-entropy loss function may alleviate this problem. After numerous comparison and ablation experiments, it is verified that the improved model can indeed effectively extract amino acid sequence feature information, alleviate overfitting, and thus improve the overall prediction accuracy.

Effect of concrete unit weight on the mechanical properties of bottom ash aggregate concrete

The depletion of natural aggregates has already progressed significantly in some regions and countries, including South Korea, and it is critical to identify replacement materials to address this problem. In recent decades, researchers have focused on low-cost industrial residues or solid waste as potential solutions, and this study involves application of bottom ash (BA), a byproduct of coal-fired electrical power plants, as fine and coarse aggregates. The application of BA to the manufacture of concrete is noteworthy in terms of economic as well as ecological aspects for preserving natural resources. In the experiment, a total of 20 concrete containing bottom ash (CCBA) mixtures classified according to W/C and the bottom ash aggregate (BAA) mixing ratios were evaluated for their workability and various mechanical properties. The results of the experiment showed that the factor having the greatest effects on the mechanical properties of concrete was the usage of coarse BAA or lack thereof, and that the rates of decline in mechanical properties due to the replacement of fine BAA were within levels that could be controlled by other design factors, such as W/C. This study provides a comparison between the experimental results and the prediction results of several existing design codes, which will serve as basic data for proposing a reliable, predictive CCBA structural design model.

Comparative In Silico Analysis of Randomly Selected Heat Shock Proteins in Caenorhabditis elegans and Photorhabdus temperata

Heat shock proteins (HSPs) such as HSP70A, HSP90 etc. (also known as Chaperons) play an important role in folding and unfolding of proteins, an assemblage of multiprotein complexes, transportation and sorting of proteins in subcellular compartments, cell cycle control, signaling pathways, protection against stress and programmed cell death. Studies have also linked heat shock proteins with a sudden rise in temperature, which can be related to anhydrobiosis in nematodes. Considering the significance of HSPs in nematodes and bacteria, the present study was designed for their in silico analysis in Caenorhabditis elegans and Photorhabdus temperata. The availability of a vast amount of sequence data generated through various bioinformatics tools, coupled with computational biology advancements, provides an ideal framework for silico gene expression and its analysis. A detailed in silico insight into these proteins include physicochemical properties, secondary structure prediction, homology modeling, and different models. The amino acid composition data were subjected to multivariate techniques, Pearson correlation, and phylogenetic analysis. In the present study, the authors characterized different HSPs according to different stability parameters and valid structures. A detailed in silico analysis of these proteins and prediction of their activity in different conditions can be very useful in both in vitro and in vivo experiments.

Protein Sequence Analysis Using the MPI Bioinformatics Toolkit.

The MPI Bioinformatics Toolkit (https://toolkit.tuebingen.mpg.de) provides interactive access to a wide range of the best-performing bioinformatics tools and databases, including the state-of-the-art protein sequence comparison methods HHblits and HHpred. The Toolkit currently includes 35 external and in-house tools, covering functionalities such as sequence similarity searching, prediction of sequence features, and sequence classification. Due to this breadth of functionality, the tight interconnection of its constituent tools, and its ease of use, the Toolkit has become an important resource for biomedical research and for teaching protein sequence analysis to students in the life sciences. In this article, we provide detailed information on utilizing the three most widely accessed tools within the Toolkit: HHpred for the detection of homologs, HHpred in conjunction with MODELLER for structure prediction and homology modeling, and CLANS for the visualization of relationships in large sequence datasets. © 2020 The Authors. Basic Protocol 1: Sequence similarity searching using HHpred Alternate Protocol: Pairwise sequence comparison using HHpred Support Protocol: Building a custom multiple sequence alignment using PSI-BLAST and forwarding it as input to HHpred Basic Protocol 2: Calculation of homology models using HHpred and MODELLER Basic Protocol 3: Cluster analysis using CLANS.

Machine learning a model for RNA structure prediction.

RNA function crucially depends on its structure. Thermodynamic models currently used for secondary structure prediction rely on computing the partition function of folding ensembles, and can thus estimate minimum free-energy structures and ensemble populations. These models sometimes fail in identifying native structures unless complemented by auxiliary experimental data. Here, we build a set of models that combine thermodynamic parameters, chemical probing data (DMS and SHAPE) and co-evolutionary data (direct coupling analysis) through a network that outputs perturbations to the ensemble free energy. Perturbations are trained to increase the ensemble populations of a representative set of known native RNA structures. In the chemical probing nodes of the network, a convolutional window combines neighboring reactivities, enlightening their structural information content and the contribution of local conformational ensembles. Regularization is used to limit overfitting and improve transferability. The most transferable model is selected through a cross-validation strategy that estimates the performance of models on systems on which they are not trained. With the selected model we obtain increased ensemble populations for native structures and more accurate predictions in an independent validation set. The flexibility of the approach allows the model to be easily retrained and adapted to incorporate arbitrary experimental information.

Longitudinal trajectories of brain age in young individuals at familial risk of mood disorder from the Scottish Bipolar Family Study.

Background: Within young individuals, mood disorder onset may be related to changes in trajectory of brain structure development. To date, however, longitudinal prospective studies remain scarce and show partly contradictory findings, with a lack of emphasis on changes at the level of global brain patterns. Cross-sectional adult studies have applied such methods and show that mood disorders are associated with accelerated brain aging. Currently, it remains unclear whether young individuals show differential brain structure aging trajectories associated with onset of mood disorder and/or presence of familial risk. Methods: Participants included young individuals (15-30 years, 53%F) from the prospective longitudinal Scottish Bipolar Family Study with and without close family history of mood disorder. All were well at time of recruitment. Implementing a structural MRI-based brain age prediction model, we globally assessed individual trajectories of age-related structural change using the difference between predicted brain age and chronological age (brain-predicted age difference (brain-PAD)) at baseline and at 2-year follow-up. Based on follow-up clinical assessment, individuals were categorised into three groups: (i) controls who remained well (C-well, n = 93), (ii) high familial risk who remained well (HR-well, n = 74) and (iii) high familial risk who developed a mood disorder (HR-MD, n = 35). Results: At baseline, brain-PAD was comparable between groups. Results showed statistically significant negative trajectories of brain-PAD between baseline and follow-up for HR-MD versus C-well ( β = -0.60, p corrected < 0.001) and HR-well ( β = -0.36, p corrected = 0.02), with a potential intermediate trajectory for HR-well ( β = -0.24 years, p corrected = 0.06). Conclusions: These preliminary findings suggest that within young individuals, onset of mood disorder and familial risk may be associated with a deceleration in brain structure aging trajectories. Extended longitudinal research will need to corroborate findings of emerging maturational lags in relation to mood disorder risk and onset.

DeepEC: Adversarial attacks against graph structure prediction models

Inspired by the practical importance of graph structured data, link prediction, one of the most frequently applied tasks on graph data, has garnered considerable attention in recent years, and they have been widely applied in item recommendation, privacy inference attack, knowledge graph completion, fraud detection, and other fields. However, recent studies show that machine learning-based intelligent systems are vulnerable to adversarial attacks, which has recently inspired much research on the security problems of machine learning in the context of computer vision, natural language processing, physical world, etc. Nonetheless, there is a lack of understanding of the vulnerability of link prediction methods in face of adversarial attacks. To unveil the weaknesses and aid in the development of robust link prediction methods, we propose a deep architecture-based adversarial attack method, called Deep Ensemble Coding, against link prediction. In particular, based on the assumption that links play different structural roles in structure organization, we propose a deep linear coding-based structure enhancement mechanism to generate adversarial examples. We also empirically investigate other adversarial attack methods for graph data, including heuristic and evolutionary perturbation methods. Based on the comprehensive experiments conducted on various real-world networks, we can conclude that the proposed adversarial attack method has satisfactory performance for link prediction. Moreover, we can observe that state-of-the-art link prediction algorithms are vulnerable to adversarial attacks and, for adversarial defense, the attack can be viewed as a robustness evaluation for the construction of robust link prediction methods.

The structural model for prediction of perceived social support based on mindfulness in patients with Ulcerative Colitis

The structural model for prediction of perceived social support based on mindfulness in patients with Ulcerative Colitis

Longitudinal trajectories of brain age in young individuals at familial risk of mood disorder from the Scottish Bipolar Family Study.

Background: Within young individuals, mood disorder onset may be related to changes in trajectory of brain structure development. To date, however, longitudinal prospective studies remain scarce and show partly contradictory findings, with a lack of emphasis on changes at the level of global brain patterns. Cross-sectional adult studies have applied such methods and show that mood disorders are associated with accelerated brain ageing. Currently, it remains unclear whether young individuals show differential brain structure aging trajectories associated with onset of mood disorder and/or presence of familial risk. Methods: Participants included young individuals (15-30 years, 53%F) from the prospective longitudinal Scottish Bipolar Family Study with and without close family history of mood disorder. All were well at time of recruitment. Implementing a structural MRI-based brain age prediction model, we globally assessed individual trajectories of age-related structural change using the difference between predicted brain age and chronological age (brain-predicted age difference (brain-PAD)) at baseline and at 2-year follow-up. Based on follow-up clinical assessment, individuals were categorised into three groups: (i) controls who remained well (C-well, n = 93), (ii) high familial risk who remained well (HR-well, n = 74) and (iii) high familial risk who developed a mood disorder (HR-MD, n = 35). Results: At baseline, brain-PAD was comparable between groups. Results showed statistically significant negative trajectories of brain-PAD between baseline and follow-up for HR-MD versus C-well ( β = -0.60, p corrected < 0.001) and HR-well ( β = -0.36, p corrected = 0.02), with a potential intermediate trajectory for HR-well ( β = -0.24 years, p corrected = 0.06). Conclusions: These preliminary findings suggest that within young individuals, onset of mood disorder and familial risk may be associated with a deceleration in brain structure aging trajectories. Extended longitudinal research will need to corroborate findings of emerging maturational lags in relation to mood disorder risk and onset.

Longitudinal trajectories of brain age in young individuals at familial risk of mood disorder from the Scottish Bipolar Family Study

Background: Within young individuals, mood disorder onset may be related to changes in trajectory of brain structure development. To date, however, longitudinal prospective studies remain scarce and show partly contradictory findings, with a lack of emphasis on changes at the level of global brain patterns. Cross-sectional adult studies have applied such methods and show that mood disorders are associated with accelerated brain aging. Currently, it remains unclear whether young individuals show differential brain structure aging trajectories associated with onset of mood disorder and/or presence of familial risk. Methods: Participants included young individuals (15-30 years, 53%F) from the prospective longitudinal Scottish Bipolar Family Study with and without close family history of mood disorder. All were well at time of recruitment. Implementing a structural MRI-based brain age prediction model, we globally assessed individual trajectories of age-related structural change using the difference between predicted brain age and chronological age (brain-predicted age difference (brain-PAD)) at baseline and at 2-year follow-up. Based on follow-up clinical assessment, individuals were categorised into three groups: (i) controls who remained well (C-well, n = 93), (ii) high familial risk who remained well (HR-well, n = 74) and (iii) high familial risk who developed a mood disorder (HR-MD, n = 35). Results: At baseline, brain-PAD was comparable between groups. Results showed statistically significant negative trajectories of brain-PAD between baseline and follow-up for HR-MD versus C-well ( β = -0.60, p corrected < 0.001) and HR-well ( β = -0.36, p corrected = 0.02), with a potential intermediate trajectory for HR-well ( β = -0.24 years, p corrected = 0.06). Conclusions: These preliminary findings suggest that within young individuals, onset of mood disorder and familial risk may be associated with a deceleration in brain structure aging trajectories. Extended longitudinal research will need to corroborate findings of emerging maturational lags in relation to mood disorder risk and onset.

Comprehensive approach in definition of the oil and gas exploration objects priority

JSC “Ukrnafta” is the main oil producing company which carries out explorative drilling in the Dnipro-Donetsk’s and PreCarpathian basins. The decrease in number of seismic studies negatively affects discovery of new potential hydrocarbon bearing structures. Reliability and productivity of seismic surveys influences productivity of explorative drillings. The article analyzes the results of seismic surveys, data from case studies and geochemical surveys to predict the oil and gas prospective areas in the Dnipro-Donetsk basin, as well as their complexation. It is shown that need to use alternative methods of exploration has objectively been increasing, in particular – use of remote methods for determining the presence of hydrocarbons (using space images). To create predictive geological structural models based on a selected set of studies are widely used modern software products for reserves and resources estimation with statistical probabilistic approaches and associated risk computation functionalities. Comprehensive study of results carried out in previous years, utilization of modern methods and software for data analysis enables to define directions of geological exploration works for the following years in order to raise its efficiency and strengthen the reserves base.

Dr AFC: drug repositioning through anti-fibrosis characteristic.

Fibrosis is a key component in the pathogenic mechanism of a variety of diseases. These diseases involving fibrosis may share common mechanisms and therapeutic targets, and therefore common intervention strategies and medicines may be applicable for these diseases. For this reason, deliberately introducing anti-fibrosis characteristics into predictive modeling may lead to more success in drug repositioning. In this study, anti-fibrosis knowledge base was first built by collecting data from multiple resources. Both structural and biological profiles were then derived from the knowledge base and used for constructing machine learning models including Structural Profile Prediction Model (SPPM) and Biological Profile Prediction Model (BPPM). Three external public data sets were employed for validation purpose and further exploration of potential repositioning drugs in wider chemical space. The resulting SPPM and BPPM models achieve area under the receiver operating characteristic curve (area under the curve) of 0.879 and 0.972 in the training set, and 0.814 and 0.874 in the testing set. Additionally, our results also demonstrate that substantial amount of multi-targeting natural products possess notable anti-fibrosis characteristics and might serve as encouraging candidates in fibrosis treatment and drug repositioning. To leverage our methodology and findings, we developed repositioning prediction platform, drug repositioning based on anti-fibrosis characteristic that is freely accessible via https://www.biosino.org/drafc.

Structured prediction models for argumentative claim parsing from text

The internet abounds with opinions expressed in text. While a number of natural language processing techniques have been proposed for opinion analysis from text, most offer only a shallow analysis without providing any insights into reasons supporting the opinions. In online discussions, however, opinions are typically expressed as arguments, consisting of a set of claims endowed with internal semantic structure amenable to deeper analysis. In this article, we introduce the task of argumentative claim parsing (ACP), which aims at extracting semantic structures of claims from argumentative text. The task is split into two subtasks: claim segmentation and claim structuring. We present a new dataset on two discussion topics with claims manually annotated for both subtasks. Inspired by structured prediction approaches, we propose a number of supervised machine learning models for the ACP task, including deep learning, chain classifier, and joint learning models. Our experiments reveal that claim segmentation is a relatively feasible task, with the best-performing model achieving up to 0.37 and 0.79 exact and lenient macro-averaged F1-score, respectively. Claim structuring, however, proved to be a more challenging task, with the best-performing models achieving at most 0.08 macro-averaged F1-score.

A single-model quality assessment method for poor quality protein structure

BackgroundQuality assessment of protein tertiary structure prediction models, in which structures of the best quality are selected from decoys, is a major challenge in protein structure prediction, and is crucial to determine a model’s utility and potential applications. Estimating the quality of a single model predicts the model’s quality based on the single model itself. In general, the Pearson correlation value of the quality assessment method increases in tandem with an increase in the quality of the model pool. However, there is no consensus regarding the best method to select a few good models from the poor quality model pool.ResultsWe introduce a novel single-model quality assessment method for poor quality models that uses simple linear combinations of six features. We perform weighted search and linear regression on a large dataset of models from the 12th Critical Assessment of Protein Structure Prediction (CASP12) and benchmark the results on CASP13 models. We demonstrate that our method achieves outstanding performance on poor quality models.ConclusionsAccording to results of poor protein structure assessment based on six features, contact prediction and relying on fewer prediction features can improve selection accuracy.

Inductive transfer learning for molecular activity prediction: Next-Gen QSAR Models with MolPMoFiT

Deep neural networks can directly learn from chemical structures without extensive, user-driven selection of descriptors in order to predict molecular properties/activities with high reliability. But these approaches typically require large training sets to learn the endpoint-specific structural features and ensure reasonable prediction accuracy. Even though large datasets are becoming the new normal in drug discovery, especially when it comes to high-throughput screening or metabolomics datasets, one should also consider smaller datasets with challenging endpoints to model and forecast. Thus, it would be highly relevant to better utilize the tremendous compendium of unlabeled compounds from publicly-available datasets for improving the model performances for the user’s particular series of compounds. In this study, we propose the Molecular Prediction Model Fine-Tuning (MolPMoFiT) approach, an effective transfer learning method based on self-supervised pre-training + task-specific fine-tuning for QSPR/QSAR modeling. A large-scale molecular structure prediction model is pre-trained using one million unlabeled molecules from ChEMBL in a self-supervised learning manner, and can then be fine-tuned on various QSPR/QSAR tasks for smaller chemical datasets with specific endpoints. Herein, the method is evaluated on four benchmark datasets (lipophilicity, FreeSolv, HIV, and blood–brain barrier penetration). The results showed the method can achieve strong performances for all four datasets compared to other state-of-the-art machine learning modeling techniques reported in the literature so far.

Structured Output Learning with Conditional Generative Flows

Traditional structured prediction models try to learn the conditional likelihood, i.e., p(y|x), to capture the relationship between the structured output y and the input features x. For many models, computing the likelihood is intractable. These models are therefore hard to train, requiring the use of surrogate objectives or variational inference to approximate likelihood. In this paper, we propose conditional Glow (c-Glow), a conditional generative flow for structured output learning. C-Glow benefits from the ability of flow-based models to compute p(y|x exactly and efficiently. Learning with c-Glow does not require a surrogate objective or performing inference during training. Once trained, we can directly and efficiently generate conditional samples. We develop a sample-based prediction method, which can use this advantage to do efficient and effective inference. In our experiments, we test c-Glow on five different tasks. C-Glow outperforms the state-of-the-art baselines in some tasks and predicts comparable outputs in the other tasks. The results show that c-Glow is versatile and is applicable to many different structured prediction problems.

Natural language processing of clinical mental health notes may add predictive value to existing suicide risk models.

This study evaluated whether natural language processing (NLP) of psychotherapy note text provides additional accuracy over and above currently used suicide prediction models. We used a cohort of Veterans Health Administration (VHA) users diagnosed with post-traumatic stress disorder (PTSD) between 2004-2013. Using a case-control design, cases (those that died by suicide during the year following diagnosis) were matched to controls (those that remained alive). After selecting conditional matches based on having shared mental health providers, we chose controls using a 5:1 nearest-neighbor propensity match based on the VHA's structured Electronic Medical Records (EMR)-based suicide prediction model. For cases, psychotherapist notes were collected from diagnosis until death. For controls, psychotherapist notes were collected from diagnosis until matched case's date of death. After ensuring similar numbers of notes, the final sample included 246 cases and 986 controls. Notes were analyzed using Sentiment Analysis and Cognition Engine, a Python-based NLP package. The output was evaluated using machine-learning algorithms. The area under the curve (AUC) was calculated to determine models' predictive accuracy. NLP derived variables offered small but significant predictive improvement (AUC = 0.58) for patients that had longer treatment duration. A small sample size limited predictive accuracy. Study identifies a novel method for measuring suicide risk over time and potentially categorizing patient subgroups with distinct risk sensitivities. Findings suggest leveraging NLP derived variables from psychotherapy notes offers an additional predictive value over and above the VHA's state-of-the-art structured EMR-based suicide prediction model. Replication with a larger non-PTSD specific sample is required.

In silico prediction of the functional and structural consequences of the non-synonymous single nucleotide polymorphism A122V in bovine CXC chemokine receptor type 1.

The current study aimed to assess whether the A122V causal polymorphism promotes alterations in the functional and structural proprieties of the CXC chemokine receptor type 1 protein (CXCR1) of cattle Bos taurus by in silico analyses. Two amino acid sequences of bovine CXCR1 was selected from database UniProtKB/Swiss-Prot: a) non-polymorphic sequence (A7KWG0) with alanine (A) at position 122, and b) polymorphic sequence harboring the A122V polymorphism, substituting alanine by valine (V) at same position. CXCR1 sequences were submitted as input to different Bioinformatics' tools to examine the effects of this polymorphism on functional and structural stabilities, to predict eventual alterations in the 3-D structural modeling, and to estimate the quality and accuracy of the predictive models. The A122V polymorphism exerted tolerable and non-deleterious effects on the polymorphic CXCR1, and the predictive structural model for polymorphic CXCR1 revealed an alpha helix spatial structure typical of a receptor transmembrane polypeptide. Although higher variations in the distances between pairs of amino acid residues at target-positions are detected in the polymorphic CXCR1 protein, more than 97% of the amino acid residues in both models were located in favored and allowed conformational regions in Ramachandran plots. Evidences has supported that the A122V polymorphism in the CXCR1 protein is associated with increased clinical mastitis incidence in dairy cows. Thus, the findings described herein prove that the replacement of the alanine by valine amino acids provokes local conformational changes in the A122V-harboring CXCR1 protein, which could directly affect its post-translational folding mechanisms and biological functionality.

Calibrating Structured Output Predictors for Natural Language Processing.

We address the problem of calibrating prediction confidence for output entities of interest in natural language processing (NLP) applications. It is important that NLP applications such as named entity recognition and question answering produce calibrated confidence scores for their predictions, especially if the applications are to be deployed in a safety-critical domain such as healthcare. However, the output space of such structured prediction models is often too large to adapt binary or multi-class calibration methods directly. In this study, we propose a general calibration scheme for output entities of interest in neural network based structured prediction models. Our proposed method can be used with any binary class calibration scheme and a neural network model. Additionally, we show that our calibration method can also be used as an uncertainty-aware, entity-specific decoding step to improve the performance of the underlying model at no additional training cost or data requirements. We show that our method outperforms current calibration techniques for named-entity-recognition, part-of-speech and question answering. We also improve our model's performance from our decoding step across several tasks and benchmark datasets. Our method improves the calibration and model performance on out-ofdomain test scenarios as well.