Dataset Information

Repositioning Drugs to the Mitochondrial Fusion Protein 2 by Three-Tunnel Deep Neural Network for Alzheimer's Disease.

ABSTRACT: Alzheimer's disease (AD) is a common neurodegenerative dementia in the elderly. Although there is no effective drug to treat AD, proteins associated with AD have been discovered in related studies. One of the proteins is mitochondrial fusion protein 2 (Mfn2), and its regulation presumably be related to AD. However, there is no specific drug for Mfn2 regulation. In this study, a three-tunnel deep neural network (3-Tunnel DNN) model is constructed and trained on the extended Davis dataset. In the prediction of drug-target binding affinity values, the accuracy of the model is up to 88.82% and the loss value is 0.172. By ranking the binding affinity values of 1,063 approved drugs and small molecular compounds in the DrugBank database, the top 15 drug molecules are recommended by the 3-Tunnel DNN model. After removing molecular weight <200 and topical drugs, a total of 11 drug molecules are selected for literature mining. The results show that six drugs have effect on AD, which are reported in references. Meanwhile, molecular docking experiments are implemented on the 11 drugs. The results show that all of the 11 drug molecules could dock with Mfn2 successfully, and 5 of them have great binding effect.

SUBMITTER: Wang X

PROVIDER: S-EPMC7917248 | biostudies-literature | 2021

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Repositioning Drugs to the Mitochondrial Fusion Protein 2 by Three-Tunnel Deep Neural Network for Alzheimer's Disease.

Wang Xun X Zhong Yue Y Ding Mao M

Frontiers in genetics 20210215

Alzheimer's disease (AD) is a common neurodegenerative dementia in the elderly. Although there is no effective drug to treat AD, proteins associated with AD have been discovered in related studies. One of the proteins is mitochondrial fusion protein 2 (Mfn2), and its regulation presumably be related to AD. However, there is no specific drug for Mfn2 regulation. In this study, a three-tunnel deep neural network (3-Tunnel DNN) model is constructed and trained on the extended Davis dataset. In the ...[more]

PMID: 33659028

Similar Datasets

Project description:BACKGROUND:Much effort has been devoted to the discovery of specific mechanisms between drugs and single targets to date. However, as biological systems maintain homeostasis at the level of functional networks robustly controlling the internal environment, such networks commonly contain multiple redundant mechanisms designed to counteract loss or perturbation of a single member of the network. As such, investigation of therapeutics that target dysregulated pathways or processes, rather than single targets, may identify agents that function at a level of the biological organization more relevant to the pathology of complex diseases such as Parkinson's Disease (PD). Genome-wide association studies (GWAS) in PD have identified common variants underlying disease susceptibility, while gene expression microarray data provide genome-wide transcriptional profiles. These genomic studies can illustrate upstream perturbations causing the dysfunction in signaling pathways and downstream biochemical mechanisms leading to the PD phenotype. We hypothesize that drugs acting at the level of a gene expression module specific to PD can overcome the lack of efficacy associated with targeting a single gene in polygenic diseases. Thus, this approach represents a promising new direction for module-based drug discovery in human diseases such as PD. RESULTS:We built a framework that integrates GWAS data with gene co-expression modules from tissues representing three brain regions-the frontal gyrus, the lateral substantia, and the medial substantia in PD patients. Using weighted gene correlation network analysis (WGCNA) software package in R, we conducted enrichment analysis of data from a GWAS of PD. This led to the identification of two over-represented PD-specific gene co-expression network modules: the Brown Module (Br) containing 449 genes and the Turquoise module (T) containing 905 genes. Further enrichment analysis identified four functional pathways within the Br module (cellular respiration, intracellular transport, energy coupled proton transport against the electrochemical gradient, and microtubule-based movement), and one functional pathway within the T module (M-phase). Next, we utilized drug-protein regulatory relationship databases (DMAP) and developed a Drug Effect Sum Score (DESS) to evaluate all candidate drugs that might restore gene expression to normal level across the Br and T modules. Among the drugs with the 12 highest DESS scores, 5 had been reported as potential treatments for PD and 6 hold potential repositioning applications. CONCLUSION:In this study, we present a systems pharmacology framework which draws on genetic data from GWAS and gene expression microarray data to reposition drugs for PD. Our innovative approach integrates gene co-expression modules with biomolecular interaction network analysis to identify network modules critical to the PD pathway and disease mechanism. We quantify the positive effects of drugs in a DESS score that is based on known drug-target activity profiles. Our results illustrate that this modular approach is promising for repositioning drugs for use in polygenic diseases such as PD, and is capable of addressing challenges of the hindered gene target in drug repositioning approaches to date.

Project description:In this article, the authors aim to maximally utilize multimodality neuroimaging and genetic data for identifying Alzheimer's disease (AD) and its prodromal status, Mild Cognitive Impairment (MCI), from normal aging subjects. Multimodality neuroimaging data such as MRI and PET provide valuable insights into brain abnormalities, while genetic data such as single nucleotide polymorphism (SNP) provide information about a patient's AD risk factors. When these data are used together, the accuracy of AD diagnosis may be improved. However, these data are heterogeneous (e.g., with different data distributions), and have different number of samples (e.g., with far less number of PET samples than the number of MRI or SNPs). Thus, learning an effective model using these data is challenging. To this end, we present a novel three-stage deep feature learning and fusion framework, where deep neural network is trained stage-wise. Each stage of the network learns feature representations for different combinations of modalities, via effective training using the maximum number of available samples. Specifically, in the first stage, we learn latent representations (i.e., high-level features) for each modality independently, so that the heterogeneity among modalities can be partially addressed, and high-level features from different modalities can be combined in the next stage. In the second stage, we learn joint latent features for each pair of modality combination by using the high-level features learned from the first stage. In the third stage, we learn the diagnostic labels by fusing the learned joint latent features from the second stage. To further increase the number of samples during training, we also use data at multiple scanning time points for each training subject in the dataset. We evaluate the proposed framework using Alzheimer's disease neuroimaging initiative (ADNI) dataset for AD diagnosis, and the experimental results show that the proposed framework outperforms other state-of-the-art methods.

Dataset Information

Repositioning Drugs to the Mitochondrial Fusion Protein 2 by Three-Tunnel Deep Neural Network for Alzheimer's Disease.

Publications

Repositioning Drugs to the Mitochondrial Fusion Protein 2 by Three-Tunnel Deep Neural Network for Alzheimer's Disease.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets