A novel graph mining approach to predict and evaluate food-drug interactions.

Food-drug interactions (FDIs) arise when nutritional dietary consumption regulates biochemical mechanisms involved in drug metabolism. This study proposes FDMine, a novel systematic framework that models the FDI problem as a homogenous graph. Our dataset consists of 788 unique approved small molecule drugs with metabolism-related drug-drug interactions and 320 unique food items, composed of 563 unique compounds. The potential number of interactions is 87,192 and 92,143 for disjoint and joint versions of the graph. We defined several similarity subnetworks comprising food-drug similarity, drug-drug similarity, and food-food similarity networks. A unique part of the graph involves encoding the food composition as a set of nodes and calculating a content contribution score. To predict new FDIs, we considered several link prediction algorithms and various performance metrics, including the precision@top (top 1%, 2%, and 5%) of the newly predicted links. The shortest path-based method has achieved a precision of 84%, 60% and 40% for the top 1%, 2% and 5% of FDIs identified, respectively. We validated the top FDIs predicted using FDMine to demonstrate its applicability, and we relate therapeutic anti-inflammatory effects of food items informed by FDIs. FDMine is publicly available to support clinicians and researchers.

Splice2Deep: An ensemble of deep convolutional neural networks for improved splice site prediction in genomic DNA.

BACKGROUND: The accurate identification of the exon/intron boundaries is critical for the correct annotation of genes with multiple exons. Donor and acceptor splice sites (SS) demarcate these boundaries. Therefore, deriving accurate computational models to predict the SS are useful for functional annotation of genes and genomes, and for finding alternative SS associated with different diseases. Although various models have been proposed for the in silico prediction of SS, improving their accuracy is required for reliable annotation. Moreover, models are often derived and tested using the same genome, providing no evidence of broad application, i.e. to other poorly studied genomes. RESULTS: With this in mind, we developed the Splice2Deep models for SS detection. Each model is an ensemble of deep convolutional neural networks. We evaluated the performance of the models based on the ability to detect SS in Homo sapiens, Oryza sativa japonica, Arabidopsis thaliana, Drosophila melanogaster, and Caenorhabditis elegans. Results demonstrate that the models efficiently detect SS in other organisms not considered during the training of the models. Compared to the state-of-the-art tools, Splice2Deep models achieved significantly reduced average error rates of 41.97% and 28.51% for acceptor and donor SS, respectively. Moreover, the Splice2Deep cross-organism validation demonstrates that models correctly identify conserved genomic elements enabling annotation of SS in new genomes by choosing the taxonomically closest model. CONCLUSIONS: The results of our study demonstrated that Splice2Deep both achieved a considerably reduced error rate compared to other state-of-the-art models and the ability to accurately recognize SS in other organisms for which the model was not trained, enabling annotation of poorly studied or newly sequenced genomes. Splice2Deep models are implemented in Python using Keras API; the models and the data are available at https://github.com/SomayahAlbaradei/Splice_Deep.git.

Splice2Deep: An ensemble of deep convolutional neural networks for improved splice site prediction in genomic DNA.

BACKGROUND: The accurate identification of the exon/intron boundaries is critical for the correct annotation of genes with multiple exons. Donor and acceptor splice sites (SS) demarcate these boundaries. Therefore, deriving accurate computational models to predict the SS are useful for functional annotation of genes and genomes, and for finding alternative SS associated with different diseases. Although various models have been proposed for the in silico prediction of SS, improving their accuracy is required for reliable annotation. Moreover, models are often derived and tested using the same genome, providing no evidence of broad application, i.e. to other poorly studied genomes. RESULTS: With this in mind, we developed the Splice2Deep models for SS detection. Each model is an ensemble of deep convolutional neural networks. We evaluated the performance of the models based on the ability to detect SS in Homo sapiens, Oryza sativa japonica, Arabidopsis thaliana, Drosophila melanogaster, and Caenorhabditis elegans. Results demonstrate that the models efficiently detect SS in other organisms not considered during the training of the models. Compared to the state-of-the-art tools, Splice2Deep models achieved significantly reduced average error rates of 41.97% and 28.51% for acceptor and donor SS, respectively. Moreover, the Splice2Deep cross-organism validation demonstrates that models correctly identify conserved genomic elements enabling annotation of SS in new genomes by choosing the taxonomically closest model. CONCLUSIONS: The results of our study demonstrated that Splice2Deep both achieved a considerably reduced error rate compared to other state-of-the-art models and the ability to accurately recognize SS in other organisms for which the model was not trained, enabling annotation of poorly studied or newly sequenced genomes. Splice2Deep models are implemented in Python using Keras API; the models and the data are available at https://github.com/SomayahAlbaradei/Splice_Deep.git.

Production of recombinant extracellular cholesterol esterase using consistently active promoters in Burkholderia stabilis.

Burkholderia stabilis FERMP-21014 produces highly active cholesterol esterase in the presence of fatty acids. To develop an overexpression system for cholesterol esterase production, we carried out RNA sequencing analyses to screen strongly active promoters in FERMP-21014. Based on gene expression consistency analysis, we selected nine genes that were consistently expressed at high levels, following which we constructed expression vectors using their promoter sequences and achieved overproduction of extracellular cholesterol esterase under fatty acid-free conditions. Of the tested promoters, the promoter of BSFP_0720, which encodes the alkyl hydroperoxide reductase subunit AhpC, resulted in the highest cholesterol esterase activity (24.3 U mL-1). This activity level was 243-fold higher than that of the wild-type strain under fatty acid-free conditions. We confirmed that cholesterol esterase was secreted without excessive accumulation within the cells. The gene expression consistency analysis will be useful to screen promoters applicable to the overexpression of other industrially important enzymes.

Hybrid model for efficient prediction of poly(A) signals in human genomic DNA.

Polyadenylation signals (PAS) are found in most protein-coding and some non-coding genes in eukaryotes. Their accurate recognition improves understanding gene regulation mechanisms and recognition of the 3'-end of transcribed gene regions where premature or alternate transcription ends may lead to various diseases. Although different methods and tools for in-silico prediction of genomic signals have been proposed, the correct identification of PAS in genomic DNA remains challenging due to a vast number of non-relevant hexamers identical to PAS hexamers. In this study, we developed a novel method for PAS recognition. The method is implemented in a hybrid PAS recognition model (HybPAS), which is based on deep neural networks (DNNs) and logistic regression models (LRMs). One of such models is developed for each of the 12 most frequent human PAS hexamers. DNN models appeared the best for eight PAS types (including the two most frequent PAS hexamers), while LRM appeared best for the remaining four PAS types. The new models use different combinations of signal processing-based, statistical, and sequence-based features as input. The results obtained on human genomic data show that HybPAS outperforms the well-tuned state-of-the-art Omni-PolyA models, reducing the classification error for different PAS hexamers by up to 57.35% for 10 out of 12 PAS types, with Omni-PolyA models being better for two PAS types. For the most frequent PAS types, 'AATAAA' and 'ATTAAA', HybPAS reduced the error rate by 35.14% and 34.48%, respectively. On average, HybPAS reduces the error by 30.29%. HybPAS is implemented partly in Python and in MATLAB available at https://github.com/EMANG-KAUST/PolyA_Prediction_LRM_DNN.

DeepGSR: an optimized deep-learning structure for the recognition of genomic signals and regions.

MOTIVATION: Recognition of different genomic signals and regions (GSRs) in DNA is crucial for understanding genome organization, gene regulation, and gene function, which in turn generate better genome and gene annotations. Although many methods have been developed to recognize GSRs, their pure computational identification remains challenging. Moreover, various GSRs usually require a specialized set of features for developing robust recognition models. Recently, deep-learning (DL) methods have been shown to generate more accurate prediction models than 'shallow' methods without the need to develop specialized features for the problems in question. Here, we explore the potential use of DL for the recognition of GSRs. RESULTS: We developed DeepGSR, an optimized DL architecture for the prediction of different types of GSRs. The performance of the DeepGSR structure is evaluated on the recognition of polyadenylation signals (PAS) and translation initiation sites (TIS) of different organisms: human, mouse, bovine and fruit fly. The results show that DeepGSR outperformed the state-of-the-art methods, reducing the classification error rate of the PAS and TIS prediction in the human genome by up to 29% and 86%, respectively. Moreover, the cross-organisms and genome-wide analyses we performed, confirmed the robustness of DeepGSR and provided new insights into the conservation of examined GSRs across species. AVAILABILITY AND IMPLEMENTATION: DeepGSR is implemented in Python using Keras API; it is available as open-source software and can be obtained at https://doi.org/10.5281/zenodo.1117159. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

DPubChem: a web tool for QSAR modeling and high-throughput virtual screening.

High-throughput screening (HTS) performs the experimental testing of a large number of chemical compounds aiming to identify those active in the considered assay. Alternatively, faster and cheaper methods of large-scale virtual screening are performed computationally through quantitative structure-activity relationship (QSAR) models. However, the vast amount of available HTS heterogeneous data and the imbalanced ratio of active to inactive compounds in an assay make this a challenging problem. Although different QSAR models have been proposed, they have certain limitations, e.g., high false positive rates, complicated user interface, and limited utilization options. Therefore, we developed DPubChem, a novel web tool for deriving QSAR models that implement the state-of-the-art machine-learning techniques to enhance the precision of the models and enable efficient analyses of experiments from PubChem BioAssay database. DPubChem also has a simple interface that provides various options to users. DPubChem predicted active compounds for 300 datasets with an average geometric mean and F1 score of 76.68% and 76.53%, respectively. Furthermore, DPubChem builds interaction networks that highlight novel predicted links between chemical compounds and biological assays. Using such a network, DPubChem successfully suggested a novel drug for the Niemann-Pick type C disease. DPubChem is freely available at www.cbrc.kaust.edu.sa/dpubchem .

HOCOMOCO: towards a complete collection of transcription factor binding models for human and mouse via large-scale ChIP-Seq analysis.

We present a major update of the HOCOMOCO collection that consists of patterns describing DNA binding specificities for human and mouse transcription factors. In this release, we profited from a nearly doubled volume of published in vivo experiments on transcription factor (TF) binding to expand the repertoire of binding models, replace low-quality models previously based on in vitro data only and cover more than a hundred TFs with previously unknown binding specificities. This was achieved by systematic motif discovery from more than five thousand ChIP-Seq experiments uniformly processed within the BioUML framework with several ChIP-Seq peak calling tools and aggregated in the GTRD database. HOCOMOCO v11 contains binding models for 453 mouse and 680 human transcription factors and includes 1302 mononucleotide and 576 dinucleotide position weight matrices, which describe primary binding preferences of each transcription factor and reliable alternative binding specificities. An interactive interface and bulk downloads are available on the web: http://hocomoco.autosome.ru and http://www.cbrc.kaust.edu.sa/hocomoco11. In this release, we complement HOCOMOCO by MoLoTool (Motif Location Toolbox, http://molotool.autosome.ru) that applies HOCOMOCO models for visualization of binding sites in short DNA sequences.

Omni-PolyA: a method and tool for accurate recognition of Poly(A) signals in human genomic DNA.

BACKGROUND: Polyadenylation is a critical stage of RNA processing during the formation of mature mRNA, and is present in most of the known eukaryote protein-coding transcripts and many long non-coding RNAs. The correct identification of poly(A) signals (PAS) not only helps to elucidate the 3'-end genomic boundaries of a transcribed DNA region and gene regulatory mechanisms but also gives insight into the multiple transcript isoforms resulting from alternative PAS. Although progress has been made in the in-silico prediction of genomic signals, the recognition of PAS in DNA genomic sequences remains a challenge. RESULTS: In this study, we analyzed human genomic DNA sequences for the 12 most common PAS variants. Our analysis has identified a set of features that helps in the recognition of true PAS, which may be involved in the regulation of the polyadenylation process. The proposed features, in combination with a recognition model, resulted in a novel method and tool, Omni-PolyA. Omni-PolyA combines several machine learning techniques such as different classifiers in a tree-like decision structure and genetic algorithms for deriving a robust classification model. We performed a comparison between results obtained by state-of-the-art methods, deep neural networks, and Omni-PolyA. Results show that Omni-PolyA significantly reduced the average classification error rate by 35.37% in the prediction of the 12 considered PAS variants relative to the state-of-the-art results. CONCLUSIONS: The results of our study demonstrate that Omni-PolyA is currently the most accurate model for the prediction of PAS in human and can serve as a useful complement to other PAS recognition methods. Omni-PolyA is publicly available as an online tool accessible at www.cbrc.kaust.edu.sa/omnipolya/ .

OmniGA: Optimized Omnivariate Decision Trees for Generalizable Classification Models.

Classification problems from different domains vary in complexity, size, and imbalance of the number of samples from different classes. Although several classification models have been proposed, selecting the right model and parameters for a given classification task to achieve good performance is not trivial. Therefore, there is a constant interest in developing novel robust and efficient models suitable for a great variety of data. Here, we propose OmniGA, a framework for the optimization of omnivariate decision trees based on a parallel genetic algorithm, coupled with deep learning structure and ensemble learning methods. The performance of the OmniGA framework is evaluated on 12 different datasets taken mainly from biomedical problems and compared with the results obtained by several robust and commonly used machine-learning models with optimized parameters. The results show that OmniGA systematically outperformed these models for all the considered datasets, reducing the F1 score error in the range from 100% to 2.25%, compared to the best performing model. This demonstrates that OmniGA produces robust models with improved performance. OmniGA code and datasets are available at www.cbrc.kaust.edu.sa/omniga/.

Dragon TIS Spotter: an Arabidopsis-derived predictor of translation initiation sites in plants.

SUMMARY: In higher eukaryotes, the identification of translation initiation sites (TISs) has been focused on finding these signals in cDNA or mRNA sequences. Using Arabidopsis thaliana (A.t.) information, we developed a prediction tool for signals within genomic sequences of plants that correspond to TISs. Our tool requires only genome sequence, not expressed sequences. Its sensitivity/specificity is for A.t. (90.75%/92.2%), for Vitis vinifera (66.8%/94.4%) and for Populus trichocarpa (81.6%/94.4%), which suggests that our tool can be used in annotation of different plant genomes. We provide a list of features used in our model. Further study of these features may improve our understanding of mechanisms of the translation initiation. AVAILABILITY AND IMPLEMENTATION: Our tool is implemented as an artificial neural network. It is available as a web-based tool and, together with the source code, the list of features, and data used for model development, is accessible at http://cbrc.kaust.edu.sa/dts.