DNdisorder: predicting protein disorder using boosting and deep networks

Analysis of Emerging Variants of Turkey Reovirus using Machine Learning

Avian reoviruses continue to cause disease in turkeys with varied pathogenicity and tissue tropism. Turkey enteric reovirus has been identified as a causative agent of enteritis or inapparent infections in turkeys. The new emerging variants of turkey reovirus, tentatively named turkey arthritis reovirus (TARV) and turkey hepatitis reovirus (THRV), are linked to tenosynovitis/arthritis and hepatitis, respectively. Turkey arthritis and hepatitis reoviruses are causing significant economic losses to the turkey industry. These infections can lead to poor weight gain, uneven growth, poor feed conversion, increased morbidity and mortality and reduced marketability of commercial turkeys. To combat these issues, detecting and classifying the types of reoviruses in turkey populations is essential. This research aims to employ clustering methods, specifically K-means and Hierarchical clustering, to differentiate three types of turkey reoviruses and identify novel emerging variants. Additionally, it focuses on classifying variants of turkey reoviruses by leveraging various machine learning algorithms such as Support Vector Machines, Naive Bayes, Random Forest, Decision Tree, and deep learning algorithms, including convolutional neural networks (CNNs). The experiments use real turkey reovirus sequence data, allowing for robust analysis and evaluation of the proposed methods. The results indicate that machine learning methods achieve an average accuracy of 92%, F1-Macro of 93% and F1-Weighted of 92% scores in classifying reovirus types. In contrast, the CNN model demonstrates an average accuracy of 85%, F1-Macro of 71% and F1-Weighted of 84% scores in the same classification task. The superior performance of the machine learning classifiers provides valuable insights into reovirus evolution and mutation, aiding in detecting emerging variants of pathogenic TARVs and THRVs.

Classification of DNA Sequence Based on a Non-gradient Algorithm: Pseudoinverse Learners

This chapter proposes a prototype-based classification approach for analyzing DNA barcodes that uses a spectral representation of DNA sequences and a non-gradient neural network. Biological sequences can be viewed as data components with higher non-fixed dimensions, which correspond to the length of the sequences. Through computational procedures such as one-hot encoding, numerical encoding plays an important role in DNA sequence evaluation (OHE). However, the OHE method has some disadvantages: (1) It does not add any details that could result in an additional predictive variable, and (2) if the variable has many classes, OHE significantly expands the feature space. To address these shortcomings, this chapter proposes a computationally efficient framework for classifying DNA sequences of living organisms in the image domain. A multilayer perceptron trained by a pseudoinverse learning autoencoder (PILAE) algorithm is used in the proposed strategy. The learning control parameters and the number of hidden layers do not have to be specified during the PILAE training process. As a result, the PILAE classifier outperforms other deep neural network (DNN) strategies such as the VGG-16 and Xception models.

DeepDRP: Prediction of intrinsically disordered regions based on integrated view deep learning architecture from transformer-enhanced and protein information

Intrinsic disorder in proteins, a widely distributed phenomenon in nature, is related to many crucial biological processes and various diseases. Traditional determination methods tend to be costly and labor-intensive, therefore it is desirable to seek an accurate identification method of intrinsically disordered proteins (IDPs). In this paper, we proposed a novel Deep learning model for Intrinsically Disordered Regions in Proteins named DeepDRP. DeepDRP employed an innovative TimeDistributed strategy and Bi-LSTM architecture to predict IDPs and is driven by integrated view features of PSSM, Energy-based encoding, AAindex, and transformer-enhanced embeddings including DR-BERT, OntoProtein, Prot-T5, and ESM-2. The comparison of different feature combinations indicates that the transformer-enhanced features contribute far more than traditional features to predict IDPs and ESM-2 accounts for a larger contribution in the pre-trained fusion vectors. The ablation test verified that the TimeDistributed strategy surely increased the model performance and is an efficient approach to the IDP prediction. Compared with eight state-of-the-art methods on the DISORDER723, S1, and DisProt832 datasets, the Matthews correlation coefficient of DeepDRP significantly outperformed competing methods by 4.90 % to 36.20 %, 11.80 % to 26.33 %, and 4.82 % to 13.55 %. In brief, DeepDRP is a reliable model for IDP prediction and is freely available at https://github.com/ZX-COLA/DeepDRP.

Tutorial: a guide for the selection of fast and accurate computational tools for the prediction of intrinsic disorder in proteins

Intrinsic disorder is instrumental for a wide range of protein functions, and its analysis, using computational predictions from primary structures, complements secondary and tertiary structure-based approaches. In this Tutorial, we provide an overview and comparison of 23 publicly available computational tools with complementary parameters useful for intrinsic disorder prediction, partly relying on results from the Critical Assessment of protein Intrinsic Disorder prediction experiment. We consider factors such as accuracy, runtime, availability and the need for functional insights. The selected tools are available as web servers and downloadable programs, offer state-of-the-art predictions and can be used in a high-throughput manner. We provide examples and instructions for the selected tools to illustrate practical aspects related to the submission, collection and interpretation of predictions, as well as the timing and their limitations. We highlight two predictors for intrinsically disordered proteins, flDPnn as accurate and fast and IUPred as very fast and moderately accurate, while suggesting ANCHOR2 and MoRFchibi as two of the best-performing predictors for intrinsically disordered region binding. We link these tools to additional resources, including databases of predictions and web servers that integrate multiple predictive methods. Altogether, this Tutorial provides a hands-on guide to comparatively evaluating multiple predictors, submitting and collecting their own predictions, and reading and interpreting results. It is suitable for experimentalists and computational biologists interested in accurately and conveniently identifying intrinsic disorder, facilitating the functional characterization of the rapidly growing collections of protein sequences.

Deep learning in prediction of intrinsic disorder in proteins

Intrinsic disorder prediction is an active area that has developed over 100 predictors. We identify and investigate a recent trend towards the development of deep neural network (DNN)-based methods. The first DNN-based method was released in 2013 and since 2019 deep learners account for majority of the new disorder predictors. We find that the 13 currently available DNN-based predictors are diverse in their topologies, sizes of their networks and the inputs that they utilize. We empirically show that the deep learners are statistically more accurate than other types of disorder predictors using the blind test dataset from the recent community assessment of intrinsic disorder predictions (CAID). We also identify several well-rounded DNN-based predictors that are accurate, fast and/or conveniently available. The popularity, favorable predictive performance and architectural flexibility suggest that deep networks are likely to fuel the development of future disordered predictors. Novel hybrid designs of deep networks could be used to adequately accommodate for diversity of types and flavors of intrinsic disorder. We also discuss scarcity of the DNN-based methods for the prediction of disordered binding regions and the need to develop more accurate methods for this prediction.

DeepOlf: Deep Neural Network Based Architecture for Predicting Odorants and Their Interacting Olfactory Receptors

Olfaction transduction mechanism is triggered by the binding of odorants to the specific olfactory receptors (OR's) present in the nasal cavity. Different odorants stimulate different OR's due to the difference in shape, physical and chemical properties. In this paper, a deep neural network architecture DeepOlf, based on molecular features and fingerprints of odorants and ORs, to predict whether a chemical compound is a potential odorant or not along with its interacting OR is proposed. Odorant identification and Odorant-OR interaction were modeled as a binary classification through multiple classifiers. The evaluation of these classifier's performance showed that the deep-neural network framework not only fits data with better accuracy in comparison to other classical methods (SVM, RF, k-NN) but also able to predict odorant-OR interactions more accurately. To our knowledge, this study is the first realization of deep learning ideas for the problem of odorant and interacting OR prediction. The accuracy of DeepOlf was found to be 94.83 and 99.92 percent for the prediction of odorants and Odorant- OR interactions respectively. Comparison of DeepOlf prediction with the existing SVM based prediction server, ODORactor, showed that better performance can be achieved with the proposed deep learning approach. The DeepOlf tool can be accessed at https://bioserver.iiita.ac.in/deepolf/.

Surveying over 100 predictors of intrinsic disorder in proteins

INTRODUCTION

Intrinsic disorder prediction field develops, assesses, and deploys computational predictors of disorder in protein sequences and constructs and disseminates databases of these predictions. Over 40 years of research resulted in the release of numerous resources.

AREAS COVERED

We identify and briefly summarize the most comprehensive to date collection of over 100 disorder predictors. We focus on their predictive models, availability and predictive performance. We categorize and study them from a historical point of view to highlight informative trends.

EXPERT OPINION

We find a consistent trend of improvements in predictive quality as newer and more advanced predictors are developed. The original focus on machine learning methods has shifted to meta-predictors in early 2010s, followed by a recent transition to deep learning. The use of deep learners will continue in foreseeable future given recent and convincing success of these methods. Moreover, a broad range of resources that facilitate convenient collection of accurate disorder predictions is available to users. They include web servers and standalone programs for disorder prediction, servers that combine prediction of disorder and disorder functions, and large databases of pre-computed predictions. We also point to the need to address the shortage of accurate methods that predict disordered binding regions.

A Survey of Deep Network Techniques All Classifiers Can Adopt

Deep neural networks (DNNs) have introduced novel and useful tools to the machine learning community. Other types of classifiers can potentially make use of these tools as well to improve their performance and generality. This paper reviews the current state of the art for deep learning classifier technologies that are being used outside of deep neural networks. Non-neural network classifiers can employ many components found in DNN architectures. In this paper, we review the feature learning, optimization, and regularization methods that form a core of deep network technologies. We then survey non-neural network learning algorithms that make innovative use of these methods to improve classification performance. Because many opportunities and challenges still exist, we discuss directions that can be pursued to expand the area of deep learning for a variety of classification algorithms.

Explainable deep neural networks for novel viral genome prediction

Viral infection causes a wide variety of human diseases including cancer and COVID-19. Viruses invade host cells and associate with host molecules, potentially disrupting the normal function of hosts that leads to fatal diseases. Novel viral genome prediction is crucial for understanding the complex viral diseases like AIDS and Ebola. While most existing computational techniques classify viral genomes, the efficiency of the classification depends solely on the structural features extracted. The state-of-the-art DNN models achieved excellent performance by automatic extraction of classification features, but the degree of model explainability is relatively poor. During model training for viral prediction, proposed CNN, CNN-LSTM based methods (EdeepVPP, EdeepVPP-hybrid) automatically extracts features. EdeepVPP also performs model interpretability in order to extract the most important patterns that cause viral genomes through learned filters. It is an interpretable CNN model that extracts vital biologically relevant patterns (features) from feature maps of viral sequences. The EdeepVPP-hybrid predictor outperforms all the existing methods by achieving 0.992 mean AUC-ROC and 0.990 AUC-PR on 19 human metagenomic contig experiment datasets using 10-fold cross-validation. We evaluate the ability of CNN filters to detect patterns across high average activation values. To further asses the robustness of EdeepVPP model, we perform leave-one-experiment-out cross-validation. It can work as a recommendation system to further analyze the raw sequences labeled as ‘unknown’ by alignment-based methods. We show that our interpretable model can extract patterns that are considered to be the most important features for predicting virus sequences through learned filters.

A deep neural network-based approach for prediction of mutagenicity of compounds

We are exposed to various chemical compounds present in the environment, cosmetics, and drugs almost every day. Mutagenicity is a valuable property that plays a significant role in establishing a chemical compound's safety. Exposure and handling of mutagenic chemicals in the environment pose a high health risk; therefore, identification and screening of these chemicals are essential. Considering the time constraints and the pressure to avoid laboratory animals' use, the shift to alternative methodologies that can establish a rapid and cost-effective detection without undue over-conservation seems critical. In this regard, computational detection and identification of the mutagens in environmental samples like drugs, pesticides, dyes, reagents, wastewater, cosmetics, and other substances is vital. From the last two decades, there have been numerous efforts to develop the prediction models for mutagenicity, and by far, machine learning methods have demonstrated some noteworthy performance and reliability. However, the accuracy of such prediction models has always been one of the major concerns for the researchers working in this area. The mutagenicity prediction models were developed using deep neural network (DNN), support vector machine, k-nearest neighbor, and random forest. The developed classifiers were based on 3039 compounds and validated on 1014 compounds; each of them encoded with 1597 molecular feature vectors. DNN-based prediction model yielded highest prediction accuracy of 92.95% and 83.81% with the training and test data, respectively. The area under the receiver's operating curve and precision-recall curve values were found to be 0.894 and 0.838, respectively. The DNN-based classifier not only fits the data with better performance as compared to traditional machine learning algorithms, viz., support vector machine, k-nearest neighbor, and random forest (with and without feature reduction) but also yields better performance metrics. In current work, we propose a DNN-based model to predict mutagenicity of compounds.

GHS-NET a generic hybridized shallow neural network for multi-label biomedical text classification

Exponential growth of biomedical literature and clinical data demands more robust yet precise computational methodologies to extract useful insights from biomedical literature and to perform accurate assignment of disease-specific codes. Such approaches can largely enhance the effectiveness of diverse biomedicine and bioinformatics applications. State-of-the-art computational biomedical text classification methodologies either solely leverage discrimintaive features extracted through convolution operations performed by deep convolutional neural network or contextual information extracted by recurrent neural network. However, none of the methodology takes advantage of both convolutional and recurrent neural networks. Further, existing methodologies lack to produce decent performance for the classification of different genre biomedical text such as biomedical literature or clinical notes. We, for the very first time, present a generic deep learning based hybrid multi-label classification methodology namely GHS-NET which can be utilized to accurately classify biomedical text of diverse genre. GHS-NET makes use of convolutional neural network to extract most discriminative features and bi-directional Long Short-Term Memory to acquire contextual information. GHS-NET effectiveness is evaluated for extreme multi-label biomedical literature classification and assignment of ICD-9 codes to clinical notes. For the task of extreme multi-label biomedical literature classification, performance comparison of GHS-Net and state-of-the-art deep learning based methodology reveals that GHS-Net marks the increment of 1%, 6%, and 1% for hallmarks of cancer dataset, 10%, 16%, and 11% for chemical exposure dataset in terms of precision, recall, and F1-score. For the task of clinical notes classification, GHS-Net outperforms previous best deep learning based methodology over Medical Information Mart for Intensive Care dataset (MIMIC-III) by the significant margin of 6%, 8% in terms of recall and F1-score. GHS-NET is available as a web service at¹ and potentially can be used to accurately classify multi-variate disease and chemical exposure specific text.

SMILES to Smell: Decoding the Structure-Odor Relationship of Chemical Compounds Using the Deep Neural Network Approach

Finding the relationship between the structure of an odorant molecule and its associated smell has always been an extremely challenging task. The major limitation in establishing the structure-odor relation is the vague and ambiguous nature of the descriptor-labeling, especially when the sources of odorant molecules are different. With the advent of deep networks, data-driven approaches have been substantiated to achieve more accurate linkages between the chemical structure and its smell. In this study, the deep neural network (DNN) with physiochemical properties and molecular fingerprints (PPMF) and the convolution neural network (CNN) with chemical-structure images (IMG) are developed to predict the smells of chemicals using their SMILES notations. A data set of 5185 chemical compounds with 104 smell percepts was used to develop the multilabel prediction models. The accuracies of smell prediction from DNN + PPMF and CNN + IMG (Xception based) were found to be 97.3 and 98.3%, respectively, when applied on an independent test set of chemicals. The deep learning architecture combining both DNN + PPMF and CNN + IMG prediction models is proposed, which classifies smells and may help understand the generic mechanism underlying the relationship between chemical structure and smell perception.

Bioimage-Based Prediction of Protein Subcellular Location in Human Tissue with Ensemble Features and Deep Networks

Prediction of protein subcellular location has currently become a hot topic because it has been proven to be useful for understanding both the disease mechanisms and novel drug design. With the rapid development of automated microscopic imaging technology in recent years, classification methods of bioimage-based protein subcellular location have attracted considerable attention for images can describe the protein distribution intuitively and in detail. In the current study, a prediction method of protein subcellular location was proposed based on multi-view image features that are extracted from three different views, including the four texture features of the original image, the global and local features of the protein extracted from the protein channel images after color segmentation, and the global features of DNA extracted from the DNA channel image. Finally, the extracted features were combined together to improve the performance of subcellular localization prediction. From the performance comparison of different combination features under the same classifier, the best ensemble features could be obtained. In this work, a classifier based on Stacked Auto-encoders and the random forest was also put forward. To improve the prediction results, the deep network was combined with the traditional statistical classification methods. Stringent cross-validation and independent validation tests on the benchmark dataset demonstrated the efficacy of the proposed method.

Predicting Protein Phosphorylation Sites Based on Deep Learning

Background:

Protein phosphorylation is one of the most important Post-translational Modifications (PTMs) occurring at amino acid residues serine (S), threonine (T), and tyrosine (Y). It plays critical roles in protein structure and function predicting. With the development of novel high-throughput sequencing technologies, there are a huge amount of protein sequences being generated and stored in databases.

Objective:

It is of great importance in both basic research and drug development to quickly and accurately predict which residues of S, T, or Y can be phosphorylated.

Methods:

In order to solve the problem, a novel hybrid deep learning model with a convolutional neural network and bi-directional long short-term memory recurrent neural network (CNN+BLSTM) is proposed for predicting phosphorylation sites in proteins. The model contains a list of layers that transform the input data into an output class, in which the convolution layer captures higher-level abstraction features of amino acid, while the recurrent layer captures long-term dependencies between amino acids to improve predictions. The joint model learns interactions between higher-level features derived from the protein sequence to predict the phosphorylated sites.

Results:

We applied our model together with two canonical methods namely iPhos-PseEn and MusiteDeep. A 5-fold cross-validation process indicated that CNN+BLSTM outperforms the two competitors in various evaluation metrics like the area under the receiver operating characteristic and precision-recall curves, the Matthews correlation coefficient, F-measure, accuracy, and so on.

Conclusion:

CNN+BLSTM is promising in identifying potential protein phosphorylation for further experimental validation.

Predicting sites of epitranscriptome modifications using unsupervised representation learning based on generative adversarial networks

Epitranscriptome is an exciting area that studies different types of modifications in transcripts and the prediction of such modification sites from the transcript sequence is of significant interest. However, the scarcity of positive sites for most modifications imposes critical challenges for training robust algorithms. To circumvent this problem, we propose MR-GAN, a generative adversarial network (GAN) based model, which is trained in an unsupervised fashion on the entire pre-mRNA sequences to learn a low dimensional embedding of transcriptomic sequences. MR-GAN was then applied to extract embeddings of the sequences in a training dataset we created for eight epitranscriptome modifications, including m⁶A, m¹A, m¹G, m²G, m⁵C, m⁵U, 2'-O-Me, Pseudouridine (Ψ) and Dihydrouridine (D), of which the positive samples are very limited. Prediction models were trained based on the embeddings extracted by MR-GAN. We compared the prediction performance with the one-hot encoding of the training sequences and SRAMP, a state-of-the-art m⁶A site prediction algorithm and demonstrated that the learned embeddings outperform one-hot encoding by a significant margin for up to 15% improvement. Using MR-GAN, we also investigated the sequence motifs for each modification type and uncovered known motifs as well as new motifs not possible with sequences directly. The results demonstrated that transcriptome features extracted using unsupervised learning could lead to high precision for predicting multiple types of epitranscriptome modifications, even when the data size is small and extremely imbalanced.

Deep learning in omics: a survey and guideline

Omics, such as genomics, transcriptome and proteomics, has been affected by the era of big data. A huge amount of high dimensional and complex structured data has made it no longer applicable for conventional machine learning algorithms. Fortunately, deep learning technology can contribute toward resolving these challenges. There is evidence that deep learning can handle omics data well and resolve omics problems. This survey aims to provide an entry-level guideline for researchers, to understand and use deep learning in order to solve omics problems. We first introduce several deep learning models and then discuss several research areas which have combined omics and deep learning in recent years. In addition, we summarize the general steps involved in using deep learning which have not yet been systematically discussed in the existent literature on this topic. Finally, we compare the features and performance of current mainstream open source deep learning frameworks and present the opportunities and challenges involved in deep learning. This survey will be a good starting point and guideline for omics researchers to understand deep learning.

Deep learning for mining protein data

The recent emergence of deep learning to characterize complex patterns of protein big data reveals its potential to address the classic challenges in the field of protein data mining. Much research has revealed the promise of deep learning as a powerful tool to transform protein big data into valuable knowledge, leading to scientific discoveries and practical solutions. In this review, we summarize recent publications on deep learning predictive approaches in the field of mining protein data. The application architectures of these methods include multilayer perceptrons, stacked autoencoders, deep belief networks, two- or three-dimensional convolutional neural networks, recurrent neural networks, graph neural networks, and complex neural networks and are described from five perspectives: residue-level prediction, sequence-level prediction, three-dimensional structural analysis, interaction prediction, and mass spectrometry data mining. The advantages and deficiencies of these architectures are presented in relation to various tasks in protein data mining. Additionally, some practical issues and their future directions are discussed, such as robust deep learning for protein noisy data, architecture optimization for specific tasks, efficient deep learning for limited protein data, multimodal deep learning for heterogeneous protein data, and interpretable deep learning for protein understanding. This review provides comprehensive perspectives on general deep learning techniques for protein data analysis.

End-to-End Representation Learning for Chemical-Chemical Interaction Prediction

Chemical-chemical interaction (CCI) plays a major role in predicting candidate drugs, toxicities, therapeutic effects, and biological functions. CCI is typically inferred from a variety of information; however, CCI has yet not been predicted using a learning-based approach. In other drug analyses, deep learning has been actively used in recent years. However, in most cases, deep learning has been used only for classification even though it has feature extraction capabilities. Thus, in this paper, we propose an end-to-end representation learning method for CCI, named DeepCCI, which includes feature extraction and a learning-based approach. Our proposed architecture is based on the Siamese network. Hidden representations are extracted from a simplified molecular input line entry system (SMILES), which is a string notation representing the chemical structure using weight-shared convolutional neural networks. Subsequently, L1 element-wise distances between the two extracted hidden representations are measured. The performance of DeepCCI is compared with those of 12 fingerprint-method combinations. The proposed DeepCCI shows the best performance in most of the evaluation metrics used. In addition, DeepCCI was experimentally validated to guarantee the commutative property. The automatically extracted features can alleviate the efforts required for manual feature engineering and improve prediction performance.

Characterization of bovine (Bos taurus) imprinted genes from genomic to amino acid attributes by data mining approaches

Genomic imprinting results in monoallelic expression of genes in mammals and flowering plants. Understanding the function of imprinted genes improves our knowledge of the regulatory processes in the genome. In this study, we have employed classification and clustering algorithms with attribute weighting to specify the unique attributes of both imprinted (monoallelic) and biallelic expressed genes. We have obtained characteristics of 22 known monoallelically expressed (imprinted) and 8 biallelic expressed genes that have been experimentally validated alongside 208 randomly selected genes in bovine (Bos taurus). Attribute weighting methods and various supervised and unsupervised algorithms in machine learning were applied. Unique characteristics were discovered and used to distinguish mono and biallelic expressed genes from each other in bovine. To obtain the accuracy of classification, 10-fold cross-validation with concerning each combination of attribute weighting (feature selection) and machine learning algorithms, was used. Our approach was able to accurately predict mono and biallelic genes using the genomics and proteomics attributes.

Deep Learning in the Biomedical Applications: Recent and Future Status

Deep neural networks represent, nowadays, the most effective machine learning technology in biomedical domain. In this domain, the different areas of interest concern the Omics (study of the genome—genomics—and proteins—transcriptomics, proteomics, and metabolomics), bioimaging (study of biological cell and tissue), medical imaging (study of the human organs by creating visual representations), BBMI (study of the brain and body machine interface) and public and medical health management (PmHM). This paper reviews the major deep learning concepts pertinent to such biomedical applications. Concise overviews are provided for the Omics and the BBMI. We end our analysis with a critical discussion, interpretation and relevant open challenges.

Using Deep Learning and Transfer Learning to Accurately Diagnose Early-Onset Glaucoma From Macular Optical Coherence Tomography Images

PURPOSE

We sought to construct and evaluate a deep learning (DL) model to diagnose early glaucoma from spectral-domain optical coherence tomography (OCT) images.

DESIGN

Artificial intelligence diagnostic tool development, evaluation, and comparison.

METHODS

This multi-institution study included pretraining data of 4316 OCT images (RS3000) from 1371 eyes with open angle glaucoma (OAG) regardless of the stage of glaucoma and 193 normal eyes. Training data included OCT-1000/2000 images from 94 eyes of 94 patients with early OAG (mean deviation > -5.0 dB) and 84 eyes of 84 normal subjects. Testing data included OCT-1000/2000 from 114 eyes of 114 patients with early OAG (mean deviation > -5.0 dB) and 82 eyes of 82 normal subjects. A DL (convolutional neural network) classifier was trained using a pretraining dataset, followed by a second round of training using an independent training dataset. The DL model input features were the 8 × 8 grid macular retinal nerve fiber layer thickness and ganglion cell complex layer thickness from spectral-domain OCT. Diagnostic accuracy was investigated in the testing dataset. For comparison, diagnostic accuracy was also evaluated using the random forests and support vector machine models. The primary outcome measure was the area under the receiver operating characteristic curve (AROC).

RESULTS

The AROC with the DL model was 93.7%. The AROC significantly decreased to between 76.6% and 78.8% without the pretraining process. Significantly smaller AROCs were obtained with random forests and support vector machine models (82.0% and 67.4%, respectively).

CONCLUSION

A DL model for glaucoma using spectral-domain OCT offers a substantive increase in diagnostic performance.

Deep Learning in Biomedical Data Science

Since the 1980s, deep learning and biomedical data have been coevolving and feeding each other. The breadth, complexity, and rapidly expanding size of biomedical data have stimulated the development of novel deep learning methods, and application of these methods to biomedical data have led to scientific discoveries and practical solutions. This overview provides technical and historical pointers to the field, and surveys current applications of deep learning to biomedical data organized around five subareas, roughly of increasing spatial scale: chemoinformatics, proteomics, genomics and transcriptomics, biomedical imaging, and health care. The black box problem of deep learning methods is also briefly discussed.

Artificial intelligence used in genome analysis studies

Next Generation Sequencing (NGS) or deep sequencing technology enables parallel reading of multiple individual DNA fragments, thereby enabling the identification of millions of base pairs in several hours. Recent research has clearly shown that machine learning technologies can efficiently analyse large sets of genomic data and help to identify novel gene functions and regulation regions. A deep artificial neural network consists of a group of artificial neurons that mimic the properties of living neurons. These mathematical models, termed Artificial Neural Networks (ANN), can be used to solve artificial intelligence engineering problems in several different technological fields (e.g., biology, genomics, proteomics, and metabolomics). In practical terms, neural networks are non-linear statistical structures that are organized as modelling tools and are used to simulate complex genomic relationships between inputs and outputs. To date, Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNN) have been demonstrated to be the best tools for improving performance in problem solving tasks within the genomic field.

Deep Learning and Its Applications in Biomedicine

Advances in biological and medical technologies have been providing us explosive volumes of biological and physiological data, such as medical images, electroencephalography, genomic and protein sequences. Learning from these data facilitates the understanding of human health and disease. Developed from artificial neural networks, deep learning-based algorithms show great promise in extracting features and learning patterns from complex data. The aim of this paper is to provide an overview of deep learning techniques and some of the state-of-the-art applications in the biomedical field. We first introduce the development of artificial neural network and deep learning. We then describe two main components of deep learning, i.e., deep learning architectures and model optimization. Subsequently, some examples are demonstrated for deep learning applications, including medical image classification, genomic sequence analysis, as well as protein structure classification and prediction. Finally, we offer our perspectives for the future directions in the field of deep learning.

RBSURFpred: Modeling protein accessible surface area in real and binary space using regularized and optimized regression

Accessible surface area (ASA) of a protein residue is an effective feature for protein structure prediction, binding region identification, fold recognition problems etc. Improving the prediction of ASA by the application of effective feature variables is a challenging but explorable task to consider, specially in the field of machine learning. Among the existing predictors of ASA, REGAd³p is a highly accurate ASA predictor which is based on regularized exact regression with polynomial kernel of degree 3. In this work, we present a new predictor RBSURFpred, which extends REGAd³p on several dimensions by incorporating 58 physicochemical, evolutionary and structural properties into 9-tuple peptides via Chou's general PseAAC, which allowed us to obtain higher accuracies in predicting both real-valued and binary ASA. We have compared RBSURFpred for both real and binary space predictions with state-of-the-art predictors, such as REGAd³p and SPIDER2. We also have carried out a rigorous analysis of the performance of RBSURFpred in terms of different amino acids and their properties, and also with biologically relevant case-studies. The performance of RBSURFpred establishes itself as a useful tool for the community.

Taxonomic Classification for Living Organisms Using Convolutional Neural Networks

Taxonomic classification has a wide-range of applications such as finding out more about evolutionary history. Compared to the estimated number of organisms that nature harbors, humanity does not have a thorough comprehension of to which specific classes they belong. The classification of living organisms can be done in many machine learning techniques. However, in this study, this is performed using convolutional neural networks. Moreover, a DNA encoding technique is incorporated in the algorithm to increase performance and avoid misclassifications. The algorithm proposed outperformed the state of the art algorithms in terms of accuracy and sensitivity, which illustrates a high potential for using it in many other applications in genome analysis.

A deep auto-encoder model for gene expression prediction

BACKGROUND

Gene expression is a key intermediate level that genotypes lead to a particular trait. Gene expression is affected by various factors including genotypes of genetic variants. With an aim of delineating the genetic impact on gene expression, we build a deep auto-encoder model to assess how good genetic variants will contribute to gene expression changes. This new deep learning model is a regression-based predictive model based on the MultiLayer Perceptron and Stacked Denoising Auto-encoder (MLP-SAE). The model is trained using a stacked denoising auto-encoder for feature selection and a multilayer perceptron framework for backpropagation. We further improve the model by introducing dropout to prevent overfitting and improve performance.

RESULTS

To demonstrate the usage of this model, we apply MLP-SAE to a real genomic datasets with genotypes and gene expression profiles measured in yeast. Our results show that the MLP-SAE model with dropout outperforms other models including Lasso, Random Forests and the MLP-SAE model without dropout. Using the MLP-SAE model with dropout, we show that gene expression quantifications predicted by the model solely based on genotypes, align well with true gene expression patterns.

CONCLUSION

We provide a deep auto-encoder model for predicting gene expression from SNP genotypes. This study demonstrates that deep learning is appropriate for tackling another genomic problem, i.e., building predictive models to understand genotypes' contribution to gene expression. With the emerging availability of richer genomic data, we anticipate that deep learning models play a bigger role in modeling and interpreting genomics.

Deep learning methods for protein torsion angle prediction

Background

Deep learning is one of the most powerful machine learning methods that has achieved the state-of-the-art performance in many domains. Since deep learning was introduced to the field of bioinformatics in 2012, it has achieved success in a number of areas such as protein residue-residue contact prediction, secondary structure prediction, and fold recognition. In this work, we developed deep learning methods to improve the prediction of torsion (dihedral) angles of proteins.

Results

We design four different deep learning architectures to predict protein torsion angles. The architectures including deep neural network (DNN) and deep restricted Boltzmann machine (DRBN), deep recurrent neural network (DRNN) and deep recurrent restricted Boltzmann machine (DReRBM) since the protein torsion angle prediction is a sequence related problem. In addition to existing protein features, two new features (predicted residue contact number and the error distribution of torsion angles extracted from sequence fragments) are used as input to each of the four deep learning architectures to predict phi and psi angles of protein backbone. The mean absolute error (MAE) of phi and psi angles predicted by DRNN, DReRBM, DRBM and DNN is about 20–21° and 29–30° on an independent dataset. The MAE of phi angle is comparable to the existing methods, but the MAE of psi angle is 29°, 2° lower than the existing methods. On the latest CASP12 targets, our methods also achieved the performance better than or comparable to a state-of-the art method.

Conclusions

Our experiment demonstrates that deep learning is a valuable method for predicting protein torsion angles. The deep recurrent network architecture performs slightly better than deep feed-forward architecture, and the predicted residue contact number and the error distribution of torsion angles extracted from sequence fragments are useful features for improving prediction accuracy.

Electronic supplementary material

The online version of this article (10.1186/s12859-017-1834-2) contains supplementary material, which is available to authorized users.

Comprehensive review of methods for prediction of intrinsic disorder and its molecular functions

Computational prediction of intrinsic disorder in protein sequences dates back to late 1970 and has flourished in the last two decades. We provide a brief historical overview, and we review over 30 recent predictors of disorder. We are the first to also cover predictors of molecular functions of disorder, including 13 methods that focus on disordered linkers and disordered protein-protein, protein-RNA, and protein-DNA binding regions. We overview their predictive models, usability, and predictive performance. We highlight newest methods and predictors that offer strong predictive performance measured based on recent comparative assessments. We conclude that the modern predictors are relatively accurate, enjoy widespread use, and many of them are fast. Their predictions are conveniently accessible to the end users, via web servers and databases that store pre-computed predictions for millions of proteins. However, research into methods that predict many not yet addressed functions of intrinsic disorder remains an outstanding challenge.

AUCpreD: proteome-level protein disorder prediction by AUC-maximized deep convolutional neural fields

MOTIVATION

Protein intrinsically disordered regions (IDRs) play an important role in many biological processes. Two key properties of IDRs are (i) the occurrence is proteome-wide and (ii) the ratio of disordered residues is about 6%, which makes it challenging to accurately predict IDRs. Most IDR prediction methods use sequence profile to improve accuracy, which prevents its application to proteome-wide prediction since it is time-consuming to generate sequence profiles. On the other hand, the methods without using sequence profile fare much worse than using sequence profile.

METHOD

This article formulates IDR prediction as a sequence labeling problem and employs a new machine learning method called Deep Convolutional Neural Fields (DeepCNF) to solve it. DeepCNF is an integration of deep convolutional neural networks (DCNN) and conditional random fields (CRF); it can model not only complex sequence-structure relationship in a hierarchical manner, but also correlation among adjacent residues. To deal with highly imbalanced order/disorder ratio, instead of training DeepCNF by widely used maximum-likelihood, we develop a novel approach to train it by maximizing area under the ROC curve (AUC), which is an unbiased measure for class-imbalanced data.

RESULTS

Our experimental results show that our IDR prediction method AUCpreD outperforms existing popular disorder predictors. More importantly, AUCpreD works very well even without sequence profile, comparing favorably to or even outperforming many methods using sequence profile. Therefore, our method works for proteome-wide disorder prediction while yielding similar or better accuracy than the others.

AVAILABILITY AND IMPLEMENTATION

http://raptorx2.uchicago.edu/StructurePropertyPred/predict/

CONTACT

wangsheng@uchicago.edu, jinboxu@gmail.com

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Accurate prediction of protein relative solvent accessibility using a balanced model

BACKGROUND

Protein relative solvent accessibility provides insight into understanding protein structure and function. Prediction of protein relative solvent accessibility is often the first stage of predicting other protein properties. Recent predictors of relative solvent accessibility discriminate against exposed regions as compared with buried regions, resulting in higher prediction accuracy associated with buried regions relative to exposed regions.

METHODS

Here, we propose a more accurate and balanced predictor of protein relative solvent accessibility. First, we collected known proteins in three subsets according to sequence length and constructed a balanced dataset after reducing redundancy within each subset. Next, we measured the performance associated with different variables and variable combinations to determine the best variable combination. Finally, a predictor called BMRSA was constructed for modelling and prediction, which used the balanced set as the training set, the position- specific scoring matrix, predicted secondary structure, buried-exposed profile, and length of a query sequence as variables, and the conditional random field as the machine-learning method.

RESULTS

BMRSA performance on test sets confirmed that our approach improved prediction accuracy relative to state-of-the-art approaches and was balanced in its comparison of buried and exposed regions. Our method is valuable when higher levels of accuracy in predicting exposed-residue states are required. The BMRSA is available at: http://cheminfo.tongji.edu.cn:8080/BMRSA/.

How disordered is my protein and what is its disorder for? A guide through the "dark side" of the protein universe

In the last 2 decades it has become increasingly evident that a large number of proteins are either fully or partially disordered. Intrinsically disordered proteins lack a stable 3D structure, are ubiquitous and fulfill essential biological functions. Their conformational heterogeneity is encoded in their amino acid sequences, thereby allowing intrinsically disordered proteins or regions to be recognized based on properties of these sequences. The identification of disordered regions facilitates the functional annotation of proteins and is instrumental for delineating boundaries of protein domains amenable to structural determination with X-ray crystallization. This article discusses a comprehensive selection of databases and methods currently employed to disseminate experimental and putative annotations of disorder, predict disorder and identify regions involved in induced folding. It also provides a set of detailed instructions that should be followed to perform computational analysis of disorder.

Estimation of Position Specific Energy as a Feature of Protein Residues from Sequence Alone for Structural Classification

A set of features computed from the primary amino acid sequence of proteins, is crucial in the process of inducing a machine learning model that is capable of accurately predicting three-dimensional protein structures. Solutions for existing protein structure prediction problems are in need of features that can capture the complexity of molecular level interactions. With a view to this, we propose a novel approach to estimate position specific estimated energy (PSEE) of a residue using contact energy and predicted relative solvent accessibility (RSA). Furthermore, we demonstrate PSEE can be reasonably estimated based on sequence information alone. PSEE is useful in identifying the structured as well as unstructured or, intrinsically disordered region of a protein by computing favorable and unfavorable energy respectively, characterized by appropriate threshold. The most intriguing finding, verified empirically, is the indication that the PSEE feature can effectively classify disorder versus ordered residues and can segregate different secondary structure type residues by computing the constituent energies. PSEE values for each amino acid strongly correlate with the hydrophobicity value of the corresponding amino acid. Further, PSEE can be used to detect the existence of critical binding regions that essentially undergo disorder-to-order transitions to perform crucial biological functions. Towards an application of disorder prediction using the PSEE feature, we have rigorously tested and found that a support vector machine model informed by a set of features including PSEE consistently outperforms a model with an identical set of features with PSEE removed. In addition, the new disorder predictor, DisPredict2, shows competitive performance in predicting protein disorder when compared with six existing disordered protein predictors.

AUC-Maximized Deep Convolutional Neural Fields for Protein Sequence Labeling

Deep Convolutional Neural Networks (DCNN) has shown excellent performance in a variety of machine learning tasks. This paper presents Deep Convolutional Neural Fields (DeepCNF), an integration of DCNN with Conditional Random Field (CRF), for sequence labeling with an imbalanced label distribution. The widely-used training methods, such as maximum-likelihood and maximum labelwise accuracy, do not work well on imbalanced data. To handle this, we present a new training algorithm called maximum-AUC for DeepCNF. That is, we train DeepCNF by directly maximizing the empirical Area Under the ROC Curve (AUC), which is an unbiased measurement for imbalanced data. To fulfill this, we formulate AUC in a pairwise ranking framework, approximate it by a polynomial function and then apply a gradient-based procedure to optimize it. Our experimental results confirm that maximum-AUC greatly outperforms the other two training methods on 8-state secondary structure prediction and disorder prediction since their label distributions are highly imbalanced and also has similar performance as the other two training methods on solvent accessibility prediction, which has three equally-distributed labels. Furthermore, our experimental results show that our AUC-trained DeepCNF models greatly outperform existing popular predictors of these three tasks. The data and software related to this paper are available at https://github.com/realbigws/DeepCNF_AUC.

Deep learning for computational biology

Technological advances in genomics and imaging have led to an explosion of molecular and cellular profiling data from large numbers of samples. This rapid increase in biological data dimension and acquisition rate is challenging conventional analysis strategies. Modern machine learning methods, such as deep learning, promise to leverage very large data sets for finding hidden structure within them, and for making accurate predictions. In this review, we discuss applications of this new breed of analysis approaches in regulatory genomics and cellular imaging. We provide background of what deep learning is, and the settings in which it can be successfully applied to derive biological insights. In addition to presenting specific applications and providing tips for practical use, we also highlight possible pitfalls and limitations to guide computational biologists when and how to make the most use of this new technology.

Detecting Preperimetric Glaucoma with Standard Automated Perimetry Using a Deep Learning Classifier

PURPOSE

To differentiate the visual fields (VFs) of preperimetric open-angle glaucoma (OAG) patients from the VFs of healthy eyes using a deep learning (DL) method.

DESIGN

Cohort study.

PARTICIPANTS

One hundred seventy-one preperimetric glaucoma VFs (PPGVFs) from 53 eyes in 51 OAG patients and 108 healthy eyes of 87 healthy participants.

METHODS

Preperimetric glaucoma VFs were defined as all VFs before a first diagnosis of manifest glaucoma (Anderson-Patella's criteria). In total, 171 PPGVFs from 53 eyes in 51 OAG patients and 108 VFs from 108 healthy eyes in 87 healthy participants were analyzed (all VFs were tested using the Humphrey Field Analyzer 30-2 program; Carl Zeiss Meditec, Dublin, CA). The 52 total deviation, mean deviation, and pattern standard deviation values were used as predictors in the DL classifier: a deep feed-forward neural network (FNN), along with other machine learning (ML) methods, including random forests (RF), gradient boosting, support vector machine, and neural network (NN). The area under the receiver operating characteristic curve (AUC) was used to evaluate the accuracy of discrimination for each method.

MAIN OUTCOME MEASURES

The AUCs obtained with each classifier method.

RESULTS

A significantly larger AUC of 92.6% (95% confidence interval [CI], 89.8%-95.4%) was obtained using the deep FNN classifier compared with all other ML methods: 79.0% (95% CI, 73.5%-84.5%) with RF, 77.6% (95% CI, 71.7%-83.5%) with gradient boosting, 71.2% (95% CI, 65.0%-77.5%), and 66.7% (95% CI, 60.1%-73.3%) with NN.

CONCLUSIONS

Preperimetric glaucoma VFs can be distinguished from healthy VFs with very high accuracy using a deep FNN classifier.

Applications of Deep Learning in Biomedicine

Increases in throughput and installed base of biomedical research equipment led to a massive accumulation of -omics data known to be highly variable, high-dimensional, and sourced from multiple often incompatible data platforms. While this data may be useful for biomarker identification and drug discovery, the bulk of it remains underutilized. Deep neural networks (DNNs) are efficient algorithms based on the use of compositional layers of neurons, with advantages well matched to the challenges -omics data presents. While achieving state-of-the-art results and even surpassing human accuracy in many challenging tasks, the adoption of deep learning in biomedicine has been comparatively slow. Here, we discuss key features of deep learning that may give this approach an edge over other machine learning methods. We then consider limitations and review a number of applications of deep learning in biomedical studies demonstrating proof of concept and practical utility.

Predicting Conformational Disorder

In the last two decades, it has become increasingly evident that a large number of proteins are either fully or partially disordered. Intrinsically disordered proteins are ubiquitous proteins that fulfill essential biological functions while lacking a stable 3D structure. Their conformational heterogeneity is encoded at the amino acid sequence level, thereby allowing intrinsically disordered proteins or regions to be recognized based on their sequence properties. The identification of disordered regions facilitates the functional annotation of proteins and is instrumental for delineating boundaries of protein domains amenable to crystallization. This chapter focuses on the methods currently employed for predicting disorder and identifying regions involved in induced folding.

Improving Protein Fold Recognition by Deep Learning Networks

For accurate recognition of protein folds, a deep learning network method (DN-Fold) was developed to predict if a given query-template protein pair belongs to the same structural fold. The input used stemmed from the protein sequence and structural features extracted from the protein pair. We evaluated the performance of DN-Fold along with 18 different methods on Lindahl’s benchmark dataset and on a large benchmark set extracted from SCOP 1.75 consisting of about one million protein pairs, at three different levels of fold recognition (i.e., protein family, superfamily, and fold) depending on the evolutionary distance between protein sequences. The correct recognition rate of ensembled DN-Fold for Top 1 predictions is 84.5%, 61.5%, and 33.6% and for Top 5 is 91.2%, 76.5%, and 60.7% at family, superfamily, and fold levels, respectively. We also evaluated the performance of single DN-Fold (DN-FoldS), which showed the comparable results at the level of family and superfamily, compared to ensemble DN-Fold. Finally, we extended the binary classification problem of fold recognition to real-value regression task, which also show a promising performance. DN-Fold is freely available through a web server at http://iris.rnet.missouri.edu/dnfold.

Deep Learning for Drug-Induced Liver Injury

Drug-induced liver injury (DILI) has been the single most frequent cause of safety-related drug marketing withdrawals for the past 50 years. Recently, deep learning (DL) has been successfully applied in many fields due to its exceptional and automatic learning ability. In this study, DILI prediction models were developed using DL architectures, and the best model trained on 475 drugs predicted an external validation set of 198 drugs with an accuracy of 86.9%, sensitivity of 82.5%, specificity of 92.9%, and area under the curve of 0.955, which is better than the performance of previously described DILI prediction models. Furthermore, with deep analysis, we also identified important molecular features that are related to DILI. Such DL models could improve the prediction of DILI risk in humans. The DL DILI prediction models are freely available at http://www.repharma.cn/DILIserver/DILI_home.php.

An Overview of Predictors for Intrinsically Disordered Proteins over 2010-2014

The sequence-structure-function paradigm of proteins has been changed by the occurrence of intrinsically disordered proteins (IDPs). Benefiting from the structural disorder, IDPs are of particular importance in biological processes like regulation and signaling. IDPs are associated with human diseases, including cancer, cardiovascular disease, neurodegenerative diseases, amyloidoses, and several other maladies. IDPs attract a high level of interest and a substantial effort has been made to develop experimental and computational methods. So far, more than 70 prediction tools have been developed since 1997, within which 17 predictors were created in the last five years. Here, we presented an overview of IDPs predictors developed during 2010-2014. We analyzed the algorithms used for IDPs prediction by these tools and we also discussed the basic concept of various prediction methods for IDPs. The comparison of prediction performance among these tools is discussed as well.

Boosting for high-dimensional two-class prediction

Background

In clinical research prediction models are used to accurately predict the outcome of the patients based on some of their characteristics. For high-dimensional prediction models (the number of variables greatly exceeds the number of samples) the choice of an appropriate classifier is crucial as it was observed that no single classification algorithm performs optimally for all types of data. Boosting was proposed as a method that combines the classification results obtained using base classifiers, where the sample weights are sequentially adjusted based on the performance in previous iterations. Generally boosting outperforms any individual classifier, but studies with high-dimensional data showed that the most standard boosting algorithm, AdaBoost.M1, cannot significantly improve the performance of its base classier. Recently other boosting algorithms were proposed (Gradient boosting, Stochastic Gradient boosting, LogitBoost); they were shown to perform better than AdaBoost.M1 but their performance was not evaluated for high-dimensional data.

Results

In this paper we use simulation studies and real gene-expression data sets to evaluate the performance of boosting algorithms when data are high-dimensional. Our results confirm that AdaBoost.M1 can perform poorly in this setting, often failing to improve the performance of its base classifier. We provide the explanation for this and propose a modification, AdaBoost.M1.ICV, which uses cross-validated estimates of the prediction errors and outperforms the original algorithm when data are high-dimensional. The use of AdaBoost.M1.ICV is advisable when the base classifier overfits the training data: the number of variables is large, the number of samples is small, and/or the difference between the classes is large. To a lesser extent also Gradient boosting suffers from similar problems. Contrary to the findings for the low-dimensional data, shrinkage does not improve the performance of Gradient boosting when data are high-dimensional, however it is beneficial for Stochastic Gradient boosting, which outperformed the other boosting algorithms in our analyses. LogitBoost suffers from overfitting and generally performs poorly.

Conclusions

The results show that boosting can substantially improve the performance of its base classifier also when data are high-dimensional. However, not all boosting algorithms perform equally well. LogitBoost, AdaBoost.M1 and Gradient boosting seem less useful for this type of data. Overall, Stochastic Gradient boosting with shrinkage and AdaBoost.M1.ICV seem to be the preferable choices for high-dimensional class-prediction.

Electronic supplementary material

The online version of this article (doi:10.1186/s12859-015-0723-9) contains supplementary material, which is available to authorized users.

DeepCNF-D: Predicting Protein Order/Disorder Regions by Weighted Deep Convolutional Neural Fields

Intrinsically disordered proteins or protein regions are involved in key biological processes including regulation of transcription, signal transduction, and alternative splicing. Accurately predicting order/disorder regions ab initio from the protein sequence is a prerequisite step for further analysis of functions and mechanisms for these disordered regions. This work presents a learning method, weighted DeepCNF (Deep Convolutional Neural Fields), to improve the accuracy of order/disorder prediction by exploiting the long-range sequential information and the interdependency between adjacent order/disorder labels and by assigning different weights for each label during training and prediction to solve the label imbalance issue. Evaluated by the CASP9 and CASP10 targets, our method obtains 0.855 and 0.898 AUC values, which are higher than the state-of-the-art single ab initio predictors.

An Overview of Practical Applications of Protein Disorder Prediction and Drive for Faster, More Accurate Predictions

Protein disordered regions are segments of a protein chain that do not adopt a stable structure. Thus far, a variety of protein disorder prediction methods have been developed and have been widely used, not only in traditional bioinformatics domains, including protein structure prediction, protein structure determination and function annotation, but also in many other biomedical fields. The relationship between intrinsically-disordered proteins and some human diseases has played a significant role in disorder prediction in disease identification and epidemiological investigations. Disordered proteins can also serve as potential targets for drug discovery with an emphasis on the disordered-to-ordered transition in the disordered binding regions, and this has led to substantial research in drug discovery or design based on protein disordered region prediction. Furthermore, protein disorder prediction has also been applied to healthcare by predicting the disease risk of mutations in patients and studying the mechanistic basis of diseases. As the applications of disorder prediction increase, so too does the need to make quick and accurate predictions. To fill this need, we also present a new approach to predict protein residue disorder using wide sequence windows that is applicable on the genomic scale.

Improving prediction of secondary structure, local backbone angles, and solvent accessible surface area of proteins by iterative deep learning

Direct prediction of protein structure from sequence is a challenging problem. An effective approach is to break it up into independent sub-problems. These sub-problems such as prediction of protein secondary structure can then be solved independently. In a previous study, we found that an iterative use of predicted secondary structure and backbone torsion angles can further improve secondary structure and torsion angle prediction. In this study, we expand the iterative features to include solvent accessible surface area and backbone angles and dihedrals based on Cα atoms. By using a deep learning neural network in three iterations, we achieved 82% accuracy for secondary structure prediction, 0.76 for the correlation coefficient between predicted and actual solvent accessible surface area, 19° and 30° for mean absolute errors of backbone φ and ψ angles, respectively, and 8° and 32° for mean absolute errors of Cα-based θ and τ angles, respectively, for an independent test dataset of 1199 proteins. The accuracy of the method is slightly lower for 72 CASP 11 targets but much higher than those of model structures from current state-of-the-art techniques. This suggests the potentially beneficial use of these predicted properties for model assessment and ranking.