Dataset Information

Prediction of metabolic reactions based on atomic and molecular properties of small-molecule compounds.

ABSTRACT:

Motivation

Our knowledge of the metabolites in cells and their reactions is far from complete as revealed by metabolomic measurements that detect many more small molecules than are documented in metabolic databases. Here, we develop an approach for predicting the reactivity of small-molecule metabolites in enzyme-catalyzed reactions that combines expert knowledge, computational chemistry and machine learning.

Results

We classified 4843 reactions documented in the KEGG database, from all six Enzyme Commission classes (EC 1-6), into 80 reaction classes, each of which is marked by a characteristic functional group transformation. Reaction centers and surrounding local structures in substrates and products of these reactions were represented using SMARTS. We found that each of the SMARTS-defined chemical substructures is widely distributed among metabolites, but only a fraction of the functional groups in these substructures are reactive. Using atomic properties of atoms in a putative reaction center and molecular properties as features, we trained support vector machine (SVM) classifiers to discriminate between functional groups that are reactive and non-reactive. Classifier accuracy was assessed by cross-validation analysis. A typical sensitivity [TP/(TP+FN)] or specificity [TN/(TN+FP)] is ?0.8. Our results suggest that metabolic reactivity of small-molecule compounds can be predicted with reasonable accuracy based on the presence of a potentially reactive functional group and the chemical features of its local environment.

Availability

The classifiers presented here can be used to predict reactions via a web site (http://cellsignaling.lanl.gov/Reactivity/). The web site is freely available.

SUBMITTER: Mu F

PROVIDER: S-EPMC3102224 | biostudies-literature | 2011 Jun

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Prediction of metabolic reactions based on atomic and molecular properties of small-molecule compounds.

Mu Fangping F Unkefer Clifford J CJ Unkefer Pat J PJ Hlavacek William S WS

Bioinformatics (Oxford, England) 20110408 11

<h4>Motivation</h4>Our knowledge of the metabolites in cells and their reactions is far from complete as revealed by metabolomic measurements that detect many more small molecules than are documented in metabolic databases. Here, we develop an approach for predicting the reactivity of small-molecule metabolites in enzyme-catalyzed reactions that combines expert knowledge, computational chemistry and machine learning.<h4>Results</h4>We classified 4843 reactions documented in the KEGG database, fr ...[more]

PMID: 21478194

Similar Datasets

Project description:Background Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia and engenders significant global health care burden. The underlying mechanisms of AF is remained to be revealed and current treatment options for AF have limitations. Besides, a detection system can help identify those at risk of developing AF and will enable personalized management. Materials and Methods In this study, we utilized the robust rank aggregation method to integrate six AF microarray datasets from the Gene Expression Omnibus database, and identified a set of differentially expressed genes between patients with AF and controls. Potential compounds were identified by mining the Connectivity Map database. Functional modules and closely-interacted clusters were identified using weighted gene co-expression network analysis and protein–protein interaction network, respectively. The overlapped hub genes were further filtered. Subsequent analyses were performed to analyze the function, biological features, and regulatory networks. Moreover, a reliable Machine Learning-based diagnostic model was constructed and visualized to clarify the diagnostic features of these genes. Results A total of 156 upregulated and 34 downregulated genes were identified, some of which had not been previously investigated. We showed that mitogen-activated protein kinase and epidermal growth factor receptor inhibitors were likely to mitigate AF based on Connectivity Map analysis. Four genes, including CXCL12, LTBP1, LOXL1, and IGFBP3, were identified as hub genes. CXCL12 was shown to play an important role in regulation of local inflammatory response and immune cell infiltration. Regulation of CXCL12 expression in AF was analyzed by constructing a transcription factor-miRNA-mRNA network. The Machine Learning-based diagnostic model generated in this study showed good efficacy and reliability. Conclusion Key genes involving in the pathogenesis of AF and potential therapeutic compounds for AF were identified. The biological features of CXCL12 in AF were investigated using integrative bioinformatics tools. The results suggested that CXCL12 might be a biomarker that could be used for distinguishing subsets of AF, and indicated that CXCL12 might be an important intermediate in the development of AF. A reliable Machine Learning-based diagnostic model was constructed. Our work improved understanding of the mechanisms of AF predisposition and progression, and identified potential therapeutic avenues for treatment of AF.

Project description:Scientists rely on high-throughput screening tools to identify promising small-molecule compounds for the development of biochemical probes and drugs. This study focuses on the identification of promiscuous bioactive compounds, which are compounds that appear active in many high-throughput screening experiments against diverse targets but are often false-positives which may not be easily developed into successful probes. These compounds can exhibit bioactivity due to nonspecific, intractable mechanisms of action and/or by interference with specific assay technology readouts. Such "frequent hitters" are now commonly identified using substructure filters, including pan assay interference compounds (PAINS). Herein, we show that mechanistic modeling of small-molecule reactivity using deep learning can improve upon PAINS filters when modeling promiscuous bioactivity in PubChem assays. Without training on high-throughput screening data, a deep learning model of small-molecule reactivity achieves a sensitivity and specificity of 18.5% and 95.5%, respectively, in identifying promiscuous bioactive compounds. This performance is similar to PAINS filters, which achieve a sensitivity of 20.3% at the same specificity. Importantly, such reactivity modeling is complementary to PAINS filters. When PAINS filters and reactivity models are combined, the resulting model outperforms either method alone, achieving a sensitivity of 24% at the same specificity. However, as a probabilistic model, the sensitivity and specificity of the deep learning model can be tuned by adjusting the threshold. Moreover, for a subset of PAINS filters, this reactivity model can help discriminate between promiscuous and nonpromiscuous bioactive compounds even among compounds matching those filters. Critically, the reactivity model provides mechanistic hypotheses for assay interference by predicting the precise atoms involved in compound reactivity. Overall, our analysis suggests that deep learning approaches to modeling promiscuous compound bioactivity may provide a complementary approach to current methods for identifying promiscuous compounds.

Dataset Information

Prediction of metabolic reactions based on atomic and molecular properties of small-molecule compounds.

Motivation

Results

Availability

Publications

Prediction of metabolic reactions based on atomic and molecular properties of small-molecule compounds.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets