Dataset Information

Combining heterogenous data for prediction of disease related and pharmacogenes.

ABSTRACT: Identifying genetic variants that affect drug response or play a role in disease is an important task for clinicians and researchers. Before individual variants can be explored efficiently for effect on drug response or disease relationships, specific candidate genes must be identified. While many methods rank candidate genes through the use of sequence features and network topology, only a few exploit the information contained in the biomedical literature. In this work, we train and test a classifier on known pharmacogenes from PharmGKB and present a classifier that predicts pharmacogenes on a genome-wide scale using only Gene Ontology annotations and simple features mined from the biomedical literature. Performance of F=0.86, AUC=0.860 is achieved. The top 10 predicted genes are analyzed. Additionally, a set of enriched pharmacogenic Gene Ontology concepts is produced.

SUBMITTER: Funk CS

PROVIDER: S-EPMC3910248 | biostudies-literature | 2014

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Combining heterogenous data for prediction of disease related and pharmacogenes.

Funk Christopher S CS Hunter Lawrence E LE Cohen K Bretonnel KB

Pacific Symposium on Biocomputing. Pacific Symposium on Biocomputing 20140101

Identifying genetic variants that affect drug response or play a role in disease is an important task for clinicians and researchers. Before individual variants can be explored efficiently for effect on drug response or disease relationships, specific candidate genes must be identified. While many methods rank candidate genes through the use of sequence features and network topology, only a few exploit the information contained in the biomedical literature. In this work, we train and test a clas ...[more]

PMID: 24297559

Similar Datasets

Project description:BackgroundAnalysis and prediction of complex traits using microbiome data combined with host genomic information is a topic of utmost interest. However, numerous questions remain to be answered: how useful can the microbiome be for complex trait prediction? Are estimates of microbiability reliable? Can the underlying biological links between the host's genome, microbiome, and phenome be recovered?MethodsHere, we address these issues by (i) developing a novel simulation strategy that uses real microbiome and genotype data as inputs, and (ii) using variance-component approaches (Bayesian Reproducing Kernel Hilbert Space (RKHS) and Bayesian variable selection methods (Bayes C)) to quantify the proportion of phenotypic variance explained by the genome and the microbiome. The proposed simulation approach can mimic genetic links between the microbiome and genotype data by a permutation procedure that retains the distributional properties of the data.ResultsUsing real genotype and rumen microbiota abundances from dairy cattle, simulation results suggest that microbiome data can significantly improve the accuracy of phenotype predictions, regardless of whether some microbiota abundances are under direct genetic control by the host or not. This improvement depends logically on the microbiome being stable over time. Overall, random-effects linear methods appear robust for variance components estimation, in spite of the typically highly leptokurtic distribution of microbiota abundances. The predictive performance of Bayes C was higher but more sensitive to the number of causative effects than RKHS. Accuracy with Bayes C depended, in part, on the number of microorganisms' taxa that influence the phenotype.ConclusionsWhile we conclude that, overall, genome-microbiome-links can be characterized using variance component estimates, we are less optimistic about the possibility of identifying the causative host genetic effects that affect microbiota abundances, which would require much larger sample sizes than are typically available for genome-microbiome-phenome studies. The R code to replicate the analyses is in https://github.com/miguelperezenciso/simubiome .

Project description:Background: Age-related macular degeneration (AMD) is the most common cause of visual impairment in the developed world. Despite some treatment options for late AMD, there is no intervention that blocks early AMD proceeding to the late and blinding forms. This is partly due to the lack of precise drug targets, despite great advances in genetics, epidemiology, and protein-protein interaction (PPI) networks proposed to be driving the disease pathology. A systems approach to narrow down PPI networks to specific protein drug targets would provide new therapeutic options. Materials and Methods: In this study we analyzed single cell RNAseq (RNA sequencing) datasets of 17 cell types present in choroidal, retinal pigment epithelium (RPE), and neural retina (NR) tissues to explore if a more granular analysis incorporating different cell types exposes more specific pathways and relationships. Furthermore, we developed a novel and systematic gene ontology database (SysGO) to explore if a subcellular classification of processes will further enhance the understanding of the pathogenesis of this complex disorder and its comorbidities with other age-related diseases. Results: We found that 57% of the AMD (risk) genes are among the top 25% expressed genes in ∼1 of the 17 choroidal/RPE/NR cell types, and 9% were among the top 1% of expressed genes. Using SysGO, we identified an enrichment of AMD genes in cell membrane and extracellular anatomical locations, and we found both functional enrichments (e.g., cell adhesion) and cell types (e.g., fibroblasts, microglia) not previously associated with AMD pathogenesis. We reconstructed PPI networks among the top expressed AMD genes for all 17 choroidal/RPE/NR cell types, which provides molecular and anatomical definitions of AMD phenotypes that can guide therapeutic approaches to target this complex disease. Conclusion: We provide mechanism-based AMD endophenotypes that can be exploited in vitro, using computational models and for drug discovery/repurposing.

Dataset Information

Combining heterogenous data for prediction of disease related and pharmacogenes.

Publications

Combining heterogenous data for prediction of disease related and pharmacogenes.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets