Dataset Information

Machine learning reveals that structural features distinguishing promiscuous and non-promiscuous compounds depend on target combinations.

ABSTRACT: Compounds with defined multi-target activity (promiscuity) play an increasingly important role in drug discovery. However, the molecular basis of multi-target activity is currently only little understood. In particular, it remains unclear whether structural features exist that generally characterize promiscuous compounds and set them apart from compounds with single-target activity. We have devised a test system using machine learning to systematically examine structural features that might characterize compounds with multi-target activity. Using this system, more than 860,000 diagnostic predictions were carried out. The analysis provided compelling evidence for the presence of structural characteristics of promiscuous compounds that were dependent on given target combinations, but not generalizable. Feature weighting and mapping identified characteristic substructures in test compounds. Taken together, these findings are relevant for the design of compounds with desired multi-target activity.

SUBMITTER: Feldmann C

PROVIDER: S-EPMC8042106 | biostudies-literature | 2021 Apr

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Machine learning reveals that structural features distinguishing promiscuous and non-promiscuous compounds depend on target combinations.

Feldmann Christian C Bajorath Jürgen J

Scientific reports 20210412 1

Compounds with defined multi-target activity (promiscuity) play an increasingly important role in drug discovery. However, the molecular basis of multi-target activity is currently only little understood. In particular, it remains unclear whether structural features exist that generally characterize promiscuous compounds and set them apart from compounds with single-target activity. We have devised a test system using machine learning to systematically examine structural features that might char ...[more]

PMID: 33846469

Similar Datasets

Project description:PurposeComputed tomography (CT) is an effective method for detecting and characterizing lung nodules in vivo. With the growing use of chest CT, the detection frequency of lung nodules is increasing. Noninvasive methods to distinguish malignant from benign nodules have the potential to decrease the clinical burden, risk, and cost involved in follow-up procedures on the large number of false-positive lesions detected. This study examined the benefit of including perinodular parenchymal features in machine learning (ML) tools for pulmonary nodule assessment.MethodsLung nodule cases with pathology confirmed diagnosis (74 malignant, 289 benign) were used to extract quantitative imaging characteristics from computed tomography scans of the nodule and perinodular parenchyma tissue. A ML tool development pipeline was employed using k-medoids clustering and information theory to determine efficient predictor sets for different amounts of parenchyma inclusion and build an artificial neural network classifier. The resulting ML tool was validated using an independent cohort (50 malignant, 50 benign).ResultsThe inclusion of parenchymal imaging features improved the performance of the ML tool over exclusively nodular features (P < 0.01). The best performing ML tool included features derived from nodule diameter-based surrounding parenchyma tissue quartile bands. We demonstrate similar high-performance values on the independent validation cohort (AUC-ROC = 0.965). A comparison using the independent validation cohort with the Fleischner pulmonary nodule follow-up guidelines demonstrated a theoretical reduction in recommended follow-up imaging and procedures.ConclusionsRadiomic features extracted from the parenchyma surrounding lung nodules contain valid signals with spatial relevance for the task of lung cancer risk classification. Through standardization of feature extraction regions from the parenchyma, ML tool validation performance of 100% sensitivity and 96% specificity was achieved.

Project description:Background: Lung adenocarcinoma is a common malignant tumor that ranks second in the world and has a high mortality rate. G protein-coupled receptors (GPCRs) have been reported to play an important role in cancer; however, G protein-coupled receptor-associated features have not been adequately investigated. Methods: In this study, GPCR-related genes were screened at single-cell and bulk transcriptome levels based on AUcell, single-sample gene set enrichment analysis (ssGSEA) and weighted gene co-expression network (WGCNA) analysis. And a new machine learning framework containing 10 machine learning algorithms and their multiple combinations was used to construct a consensus G protein-coupled receptor-related signature (GPCRRS). GPCRRS was validated in the training set and external validation set. We constructed GPCRRS-integrated nomogram clinical prognosis prediction tools. Multi-omics analyses included genomics, single-cell transcriptomics, and bulk transcriptomics to gain a more comprehensive understanding of prognostic features. We assessed the response of risk subgroups to immunotherapy and screened for personalized drugs targeting specific risk subgroups. Finally, the expression of key GPCRRS genes was verified by RT-qPCR. Results: In this study, we identified 10 GPCR-associated genes that were significantly associated with the prognosis of lung adenocarcinoma by single-cell transcriptome and bulk transcriptome. Univariate and multivariate showed that the survival rate was higher in low risk than in high risk, which also suggested that the model was an independent prognostic factor for LUAD. In addition, we observed significant differences in biological function, mutational landscape, and immune cell infiltration in the tumor microenvironment between high and low risk groups. Notably, immunotherapy was also relevant in the high and low risk groups. In addition, potential drugs targeting specific risk subgroups were identified. Conclusion: In this study, we constructed and validated a lung adenocarcinoma G protein-coupled receptor-related signature, which has an important role in predicting the prognosis of lung adenocarcinoma and the effect of immunotherapy. It is hypothesized that LDHA, GPX3 and DOCK4 are new potential targets for lung adenocarcinoma, which can achieve breakthroughs in prognosis prediction, targeted prevention and treatment of lung adenocarcinoma and provide important guidance for anti-tumor.

Dataset Information

Machine learning reveals that structural features distinguishing promiscuous and non-promiscuous compounds depend on target combinations.

Publications

Machine learning reveals that structural features distinguishing promiscuous and non-promiscuous compounds depend on target combinations.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets