Browse
Submit Data
Databases
API
Help

Dataset Information

17 Views

0 Connections

0 Citations

0 Reanalyses

0 Downloads

Omics score: 0

ParsVNN: parsimony visible neural networks for uncovering cancer-specific and drug-sensitive genes and pathways

ABSTRACT: Abstract Prediction of cancer-specific drug responses as well as identification of the corresponding drug-sensitive genes and pathways remains a major biological and clinical challenge. Deep learning models hold immense promise for better drug response predictions, but most of them cannot provide biological and clinical interpretability. Visible neural network (VNN) models have emerged to solve the problem by giving neurons biological meanings and directly casting biological networks into the models. However, the biological networks used in VNNs are often redundant and contain components that are irrelevant to the downstream predictions. Therefore, the VNNs using these redundant biological networks are overparameterized, which significantly limits VNNs’ predictive and explanatory power. To overcome the problem, we treat the edges and nodes in biological networks used in VNNs as features and develop a sparse learning framework ParsVNN to learn parsimony VNNs with only edges and nodes that contribute the most to the prediction task. We applied ParsVNN to build cancer-specific VNN models to predict drug response for five different cancer types. We demonstrated that the parsimony VNNs built by ParsVNN are superior to other state-of-the-art methods in terms of prediction performance and identification of cancer driver genes. Furthermore, we found that the pathways selected by ParsVNN have great potential to predict clinical outcomes as well as recommend synergistic drug combinations.

SUBMITTER: Huang X

PROVIDER: S-EPMC8557386 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Json Xml

Similar Datasets

Bayesian Neural Networks for Selection of Drug Sensitive Genes.

Project description:Recent advances in high-throughput biotechnologies have provided an unprecedented opportunity for biomarker discovery, which, from a statistical point of view, can be cast as a variable selection problem. This problem is challenging due to the high-dimensional and non-linear nature of omics data and, in general, it suffers three difficulties: (i) an unknown functional form of the nonlinear system, (ii) variable selection consistency, and (iii) high-demanding computation. To circumvent the first difficulty, we employ a feed-forward neural network to approximate the unknown nonlinear function motivated by its universal approximation ability. To circumvent the second difficulty, we conduct structure selection for the neural network, which induces variable selection, by choosing appropriate prior distributions that lead to the consistency of variable selection. To circumvent the third difficulty, we implement the population stochastic approximation Monte Carlo algorithm, a parallel adaptive Markov Chain Monte Carlo (MCMC) algorithm, on the OpenMP platform which provides a linear speedup for the simulation with the number of cores of the computer. The numerical results indicate that the proposed method can work very well for identification of relevant variables for high-dimensional nonlinear systems. The proposed method is successfully applied to identification of the genes that are associated with anticancerdrug sensitivities based on the data collected in the cancer cell line encyclopedia (CCLE) study.

| S-EPMC6660200 | biostudies-literature

Uncovering New Drug Properties in Target-Based Drug-Drug Similarity Networks.

Project description:Despite recent advances in bioinformatics, systems biology, and machine learning, the accurate prediction of drug properties remains an open problem. Indeed, because the biological environment is a complex system, the traditional approach-based on knowledge about the chemical structures-can not fully explain the nature of interactions between drugs and biological targets. Consequently, in this paper, we propose an unsupervised machine learning approach that uses the information we know about drug-target interactions to infer drug properties. To this end, we define drug similarity based on drug-target interactions and build a weighted Drug-Drug Similarity Network according to the drug-drug similarity relationships. Using an energy-model network layout, we generate drug communities associated with specific, dominant drug properties. DrugBank confirms the properties of 59.52% of the drugs in these communities, and 26.98% are existing drug repositioning hints we reconstruct with our DDSN approach. The remaining 13.49% of the drugs seem not to match the dominant pharmacologic property; thus, we consider them potential drug repurposing hints. The resources required to test all these repurposing hints are considerable. Therefore we introduce a mechanism of prioritization based on the betweenness/degree node centrality. Using betweenness/degree as an indicator of drug repurposing potential, we select Azelaic acid and Meprobamate as a possible antineoplastic and antifungal, respectively. Finally, we use a test procedure based on molecular docking to analyze Azelaic acid and Meprobamate's repurposing.

| S-EPMC7557376 | biostudies-literature

Uncovering potential Drug Targets for Tuberculosis using Protein Networks.

Project description:The emergence of HIV-TB co-infection and multi-drug resistant strains of Mycobacterium tuberculosis (Mtb) drive the need for new therapeutics against the infectious disease tuberculosis. Among the reported putative TB targets in the literature, the identification and characterization of the most probable therapeutic targets that influence the complex infectious disease, primarily through interactions with other influenced proteins, remains a statistical and computational challenge in proteomic epidemiology. Protein interaction network analysis provides an effective way to understand the relationships between protein products of genes by interconnecting networks of essential genes and its protein-protein interactions for 5 broad functional categories in Mtb. We also investigated the substructure of the protein interaction network and focused on highly connected nodes known as cliques by giving weight to the edges using data mining algorithms. Cliques containing Sulphate assimilation and Shikimate pathway enzymes appeared continuously inspite of increasing constraints applied by the K-Core algorithm during Network Decomposition. The potential target narrowed down through Systems approaches was Prephanate Dehydratase present in the Shikimate pathway this gives an insight to develop novel potential inhibitors through Structure Based Drug Design with natural compounds.

| S-EPMC3374368 | biostudies-literature

Uncovering global SUMOylation signaling networks in a site-specific manner.

Project description:SUMOylation is a reversible post-translational modification essential for genome stability. Using high-resolution MS, we have studied global SUMOylation in human cells in a site-specific manner, identifying a total of >4,300 SUMOylation sites in >1,600 proteins. To our knowledge, this is the first time that >1,000 SUMOylation sites have been identified under standard growth conditions. We quantitatively studied SUMOylation dynamics in response to SUMO protease inhibition, proteasome inhibition and heat shock. Many SUMOylated lysines have previously been reported to be ubiquitinated, acetylated or methylated, thus indicating cross-talk between SUMO and other post-translational modifications. We identified 70 phosphorylation and four acetylation events in proximity to SUMOylation sites, and we provide evidence for acetylation-dependent SUMOylation of endogenous histone H3. SUMOylation regulates target proteins involved in all nuclear processes including transcription, DNA repair, chromatin remodeling, precursor-mRNA splicing and ribosome assembly.

| S-EPMC4259010 | biostudies-literature

Drug-Drug Interaction Extraction via Convolutional Neural Networks.

Project description:Drug-drug interaction (DDI) extraction as a typical relation extraction task in natural language processing (NLP) has always attracted great attention. Most state-of-the-art DDI extraction systems are based on support vector machines (SVM) with a large number of manually defined features. Recently, convolutional neural networks (CNN), a robust machine learning method which almost does not need manually defined features, has exhibited great potential for many NLP tasks. It is worth employing CNN for DDI extraction, which has never been investigated. We proposed a CNN-based method for DDI extraction. Experiments conducted on the 2013 DDIExtraction challenge corpus demonstrate that CNN is a good choice for DDI extraction. The CNN-based DDI extraction method achieves an F-score of 69.75%, which outperforms the existing best performing method by 2.75%.

| S-EPMC4752975 | biostudies-literature

Genome-Scale Networks Link Neurodegenerative Disease Genes to ?-Synuclein through Specific Molecular Pathways.

Project description:Numerous genes and molecular pathways are implicated in neurodegenerative proteinopathies, but their inter-relationships are poorly understood. We systematically mapped molecular pathways underlying the toxicity of alpha-synuclein (?-syn), a protein central to Parkinson's disease. Genome-wide screens in yeast identified 332 genes that impact ?-syn toxicity. To "humanize" this molecular network, we developed a computational method, TransposeNet. This integrates a Steiner prize-collecting approach with homology assignment through sequence, structure, and interaction topology. TransposeNet linked ?-syn to multiple parkinsonism genes and druggable targets through perturbed protein trafficking and ER quality control as well as mRNA metabolism and translation. A calcium signaling hub linked these processes to perturbed mitochondrial quality control and function, metal ion transport, transcriptional regulation, and signal transduction. Parkinsonism gene interaction profiles spatially opposed in the network (ATP13A2/PARK9 and VPS35/PARK17) were highly distinct, and network relationships for specific genes (LRRK2/PARK8, ATXN2, and EIF4G1/PARK18) were confirmed in patient induced pluripotent stem cell (iPSC)-derived neurons. This cross-species platform connected diverse neurodegenerative genes to proteinopathy through specific mechanisms and may facilitate patient stratification for targeted therapy.

| S-EPMC5388136 | biostudies-literature

Hereditary Spastic Paraplegia: From Genes, Cells and Networks to Novel Pathways for Drug Discovery.

Project description:Hereditary spastic paraplegia (HSP) is a diverse group of Mendelian genetic disorders affecting the upper motor neurons, specifically degeneration of their distal axons in the corticospinal tract. Currently, there are 80 genes or genomic loci (genomic regions for which the causative gene has not been identified) associated with HSP diagnosis. HSP is therefore genetically very heterogeneous. Finding treatments for the HSPs is a daunting task: a rare disease made rarer by so many causative genes and many potential mutations in those genes in individual patients. Personalized medicine through genetic correction may be possible, but impractical as a generalized treatment strategy. The ideal treatments would be small molecules that are effective for people with different causative mutations. This requires identification of disease-associated cell dysfunctions shared across genotypes despite the large number of HSP genes that suggest a wide diversity of molecular and cellular mechanisms. This review highlights the shared dysfunctional phenotypes in patient-derived cells from patients with different causative mutations and uses bioinformatic analyses of the HSP genes to identify novel cell functions as potential targets for future drug treatments for multiple genotypes.

| S-EPMC8004882 | biostudies-literature

Things fall apart: biological species form unconnected parsimony networks.

Project description:The generality of operational species definitions is limited by problematic definitions of between-species divergence. A recent phylogenetic species concept based on a simple objective measure of statistically significant genetic differentiation uses between-species application of statistical parsimony networks that are typically used for population genetic analysis within species. Here we review recent phylogeographic studies and reanalyse several mtDNA barcoding studies using this method. We found that (i) alignments of DNA sequences typically fall apart into a separate subnetwork for each Linnean species (but with a higher rate of true positives for mtDNA data) and (ii) DNA sequences from single species typically stick together in a single haplotype network. Departures from these patterns are usually consistent with hybridization or cryptic species diversity.

| S-EPMC2391196 | biostudies-literature

Uncovering the Subtype-Specific Molecular Characteristics of Breast Cancer by Multiomics Analysis of Prognosis-Associated Genes, Driver Genes, Signaling Pathways, and Immune Activity.

Project description:Breast cancer is a heterogeneous malignant disease with different prognoses and has been divided into four molecular subtypes. It is believed that molecular events occurring in breast stem/progenitor cells contribute to the carcinogenesis and development of different breast cancer subtypes. However, these subtype-specific molecular characteristics are largely unknown. In this study, we employed 1217 breast cancer samples from The Cancer Genome Atlas (TCGA) database for a multiomics analysis of the molecular characteristics of different breast cancer subtypes based on PAM50 algorithms. We detected the expression changes of subtype-specific genes and revealed that the expression of particular subtype-specific genes significantly affected prognosis. We also investigated the mutations and copy number variations (CNVs) of breast cancer driver genes and the representative genes of ten signaling pathways in different subtypes and revealed several subtype-specifically altered genes. Moreover, we detected the infiltration of various immune cells in different subtypes of breast cancer and showed that the infiltration levels of major immune cell types are different among these subtypes. Additionally, we investigated the factors affecting the immune infiltration level and the immune cytolytic activity in different breast cancer subtypes, namely, the mutation burden, genome instability and cancer-associated fibroblast (CAF) infiltration. This study may shed light on the molecular events contributing to carcinogenesis and development and provide potential markers and targets for the clinical diagnosis and treatment of different breast cancer subtypes.

| S-EPMC8280810 | biostudies-literature

Statistical parsimony networks and species assemblages in Cephalotrichid nemerteans (nemertea).

Project description:BackgroundIt has been suggested that statistical parsimony network analysis could be used to get an indication of species represented in a set of nucleotide data, and the approach has been used to discuss species boundaries in some taxa.Methodology/principal findingsBased on 635 base pairs of the mitochondrial protein-coding gene cytochrome c oxidase I (COI), we analyzed 152 nemertean specimens using statistical parsimony network analysis with the connection probability set to 95%. The analysis revealed 15 distinct networks together with seven singletons. Statistical parsimony yielded three networks supporting the species status of Cephalothrix rufifrons, C. major and C. spiralis as they currently have been delineated by morphological characters and geographical location. Many other networks contained haplotypes from nearby geographical locations. Cladistic structure by maximum likelihood analysis overall supported the network analysis, but indicated a false positive result where subnetworks should have been connected into one network/species. This probably is caused by undersampling of the intraspecific haplotype diversity.Conclusions/significanceStatistical parsimony network analysis provides a rapid and useful tool for detecting possible undescribed/cryptic species among cephalotrichid nemerteans based on COI gene. It should be combined with phylogenetic analysis to get indications of false positive results, i.e., subnetworks that would have been connected with more extensive haplotype sampling.

| S-EPMC2943479 | biostudies-literature

OmicsDI is part of the ELIXIR infrastructure

OmicsDI is an Elixir interoperability service. Learn more ›

Tweets

OmicsDI Databases

PRIDE
PeptideAtlas
MassIVE
JPOST Repository
Physiome Model Repository

EGA
EVA
ENA
LINCS
PAXDB
Cell Collective

MetaboLights
Metabolomics Workbench
MetabolomeExpress
GNPS
BioModels
FAIRDOMHub

ArrayExpress
dbGaP
ExpressionAtlas
GEO
NODE

Information

Databases
Help
API
Contact us
Code on GitHub
Terms of use
Submit Data