DKD_CaseStudy_SemNet2

Code for diabetic kidney disease (DKD) case study using SemNet 2.0.

Overview

This DKD case study was performed with SemNet 2.0 to identify pathophysiological processes associated with DKD through extensive biomedical text-mining.

Authors

Krutika Patidar^a, Jennifer H. Deng^b, Cassie S. Mitchell^b,c, Ashlee N. Ford Versypt^a,d,e

^aDepartment of Chemical and Biological Engineering, University at Buffalo, The State University of New York, Buffalo, NY, USA
^bDepartment of Biomedical Engineering, Georgia Institute of Technology and Emory University School of Medicine, Atlanta, GA, USA
^cCenter for Machine Learning at Georgia Tech, Georgia Institute of Technology, Atlanta, GA, USA
^dDepartment of Biomedical Engineering, University at Buffalo, The State University of New York, Buffalo, NY, USA
^eInstitute for Artificial Intelligence and Data Science, University at Buffalo, The State University of New York, Buffalo, NY, USA

Script

• DKD-SemNet-visualization.ipynb This is a Jupyter notebook (Python 3) used to visualize data provided in the following data files.

Data files

• TOP10_combined_triple.csv This file contains the top 10% of source nodes associated with each pairwise domain (DKD, IR, GEC, KD, DB). Each source node has a name, semantic type (aapp or gngm), domain, mean HeteSim score, and count.
• MGI_triple_aapp.csv This file contains mapped ontologies of the top 10% proteins (aapp) at the intersection of DKD, IR, and GEC domains.
• MGI_triple_gngm.csv This file contains mapped ontologies of the top 10% genes (gngm) at the intersection of DKD, IR, and GEC domains.

Raw Data

• combined.csv is a full list of source nodes from the cross-domain results merged in a file.
• combined_10percentile.xlsx comprise source nodes and respective mean HeteSim scores in the top 10% across five domains (combined) and intersections of DKD, IR, and GEC domains (triple)
• search_queries_target_nodes.csv is a full list of search queries (sets of 4 target nodes) used to create the Hub networks in SemNet 2.0.
• User-SpecifiedTargetNodes.csv is a full list of user-specified target nodes that may or may not have been used in the SemNet 2.0 analysis.
• SourceNameDefinition.csv provides definitions for source nodes (proteins or gene) mentioned in the study.

Cytoscape session

• interaction-network-session.cys This saved Cytoscape session can be run using the Cytoscape (v3.10.0) software to generate the protein-protein (or protein-gene) interaction network. "PPI_network_paper" network in the session generates the PPI in the manuscript.
• interaction.sif This file contains interaction rules between proteins/genes.
• edge_rule.sif This file maps the positive (+1) or negative (-1) regulatory relation between interacting proteins as provided in interaction.sif.

Recommended Supplementary Packages

• SemNet 2.0 is a group of modules for working with semantic networks, specifically those built using the Semantic MEDLINE Database (SemMedDB) repository of semantic predications.
• Cytoscape is an open-source software platform for visualizing complex networks and integrating these with any type of attribute data.

Acknowledgements

This work was supported by National Institutes of Health grant R35GM133763 to A.N.F.V., National Science Foundation CAREER grant 2133411 to A.N.F.V., National Science Foundation CAREER grant 1944247 to C.S.M., National Institute of Health grant U19-AG056169 sub-award to C.S.M., and by the Chan Zuckerberg Initiative under grant 253558 to C.S.M.