Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Filters applied. Clear all
. 2016 Nov 15;32(22):3435-3443.
doi: 10.1093/bioinformatics/btw510. Epub 2016 Aug 2.

Predicting G Protein-Coupled Receptor Downstream Signaling by Tissue Expression

Affiliations
Free PMC article

Predicting G Protein-Coupled Receptor Downstream Signaling by Tissue Expression

Yun Hao et al. Bioinformatics. .
Free PMC article

Abstract

Motivation: G protein-coupled receptors (GPCRs) are central to how cells respond to their environment and a major class of pharmacological targets. However, comprehensive knowledge of which pathways are activated and deactivated by these essential sensors is largely unknown. To better understand the mechanism of GPCR signaling system, we integrated five independent genome-wide expression datasets, representing 275 human tissues and cell lines, with protein-protein interactions and functional pathway data.

Results: We found that tissue-specificity plays a crucial part in the function of GPCR signaling system. Only a few GPCRs are expressed in each tissue, which are coupled by different combinations of G-proteins or β-arrestins to trigger specific downstream pathways. Based on this finding, we predicted the downstream pathways of GPCR in human tissues and validated our results with L1000 knockdown data. In total, we identified 154,988 connections between 294 GPCRs and 690 pathways in 240 tissues and cell types.

Availability and implementation: The source code and results supporting the conclusions of this article are available at http://tatonettilab.org/resources/GOTE/source_code/ CONTACT: nick.tatonetti@columbia.eduSupplementary information: Supplementary data are available at Bioinformatics online.

Figures

Fig. 1.
Fig. 1.
Workflow and data used in GOTE. The left side shows the workflow of GOTE. The center gives a simple description of each step. The right side shows the data used in each step. In the first step we used tissue expression data to find specifically expressed GPCR in each tissue. Second, for each transducer (G-protein or β-arrestin), we obtained a list of binding proteins using the BioGRID PPI data. Third, the list of binding proteins is filtered by tissue expression, resulting in lists of tissue-specific binding proteins. Fourth, pathway enrichment analysis is performed based on the tissue-specific binding proteins of each transducer using Fisher’s Exact Test. For each pathway, a Z-score is calculated for each transducer. Fifth, for each pathway, the Z-scores of all G-proteins are combined together using Stouffer’s Method. The weight of each Z-score is in proportion to the tissue expression of each G-protein. The same analysis is repeated for the Z-scores of all β-arrestins. Finally, pathways with significant Z-scores are connected to GPCRs expressed in the same tissue as G-protein dependent pathways (those which are associated with G-proteins) or G-protein independent pathways (those which are associated with β-arrestins)
Fig. 2.
Fig. 2.
Results of HPM_PRT dataset. (A,C) Bar plot showing the number of predicted G-protein dependent (A) and independent (C) pathways, respectively. The x-axis indicates different GPCRs, which are grouped by six families. Each bar with a unique color indicates the number of predicted pathways in a tissue type. (B) A heatmap shows a mapping between GPCR and the Anatomical Therapeutic Chemical (ATC) Classification System categories of the drugs known to bind them. The first row indicates whether or not the GPCR is a known drug target. Each cell colored from white to deep purple indicates the percentage of drugs that target the GPCRs that are classified into each category. (D) Correlation between ATC classification and the results of GOTE. The purple cell of top row shows the GPCR targeted by drugs belonging to the Nervous category of ATC classification. The purple cell of bottom row shows the GPCRs which are connected to nervous system by GOTE (r = 0.543, P-value = 1.76e−10
Fig. 3.
Fig. 3.
Barplot of Jaccard similarity. GDP: G-protein dependent pathways. GIP: G-protein independent pathways. (A) Comparison of GOTE’s results between different cell lines in NCI60 dataset. Each bar indicates the mean pairwise Jaccard similarity of cell lines belonging to same or different cell type. The error bar indicates 95% confidence interval of mean value calculated by bootstrap. For both G-protein dependent pathways (green) and G-protein independent pathways (pink), the Jaccard similarity between same cell types is significantly higher than Jaccard similarity between different cell types. (B) A heatmap showing the pairwise jaccard similariy of G-protein independent pathways among tissues in U133A dataset. Each column or row indicates a tissue. The color of each cell is proportion to the Jaccard similarity between the column and row. Two clusters of tissues are highlighted in the heatmap: green cluster of neural tissues at the bottom and red cluster of immune tissues in the middle. (C) Comparison of GOTE’s results among four different datasets: U133A, HPM_RNA, HPM_PRT and GTEx. Each bar indicates the mean pairwise jaccard similarity of same tissue from two datasets. The error bar indicates 95% confidence interval of mean value calculated by bootstrap
Fig. 4.
Fig. 4.
Influence of four parameters on the prediction results of GOTE. GDP: G-protein dependent pathways. GIP: G-protein independent pathways. In each of four line graphs, the x-axis indicates the threshold (or 1/threshold in B) of one parameter. The y-axis indicates how much GOTE outperforms HighExp in precision when setting the parameter to the threshold of x-axis (the other three parameters as default setting). The four parameters are: (A) t1: the P-value threshold for highly expressed GPCRs. (B) t2: the GPCR specificity score threshold. (C) t3: the P-value threshold for highly expressed binding proteins of transducers. (D) t4: the P-value threshold for enriched pathways

Similar articles

See all similar articles

Cited by 4 articles

Substances

Feedback