gms | German Medical Science

Introduction: Comorbid diseases are one or more additional diseases co-occuring in the presence of a primary disease. The urgency of this issue is determined by the need to establish the molecular mechanisms for the development of complex multifactorial diseases (MDs) [1]. Most MDs in humans are complex and their development is based on the interaction between genetic and environmental factors [2]. Previous research suggests that drug interactions and contraindications represent an important aspect of comorbid diseases, especially in polypharmacy [3].

In a previous cooperation, the comorbid diseases asthma and hypertension were analysed. Key molecular factors like IL10 were identified and experimentally confirmed. Data integration efforts produced the first GenCoNet version, linking these diseases to their molecular factors and drugs targeting them.??????

Having demonstrated the ability to find relevant molecular factors and drugs, the goal for an updated GenCoNet version is the addition of more diseases. This allows the analysis of the GenCoNet graph for any disease pair for which sufficient data is available, like type 2 diabetes mellitus and asthma [4].??????

Implementation: The first GenCoNet version was built as a semi-automated workflow focused on two diseases. For more information to be integrated, a new pipeline was developed, automating the graph database generation from heterogeneous sources. Previously data was filtered for asthma and hypertension. Now the same information are integrated for all diseases available. Additionally, new sources were selected and integrated, to find and strengthen relationships and to introduce new entities into the graph:

• Gene Ontology terms were added to annotate genes with functional terms for clustering
• Adverse drug reaction annotations from DrugBank and PharmGKB can help in understanding relationships between drugs and diseases in the context of comorbid diseases.
• Non-coding RNAs are known to regulate gene expression, other RNAs and more. The addition of RNA interactions may further advise the analysis of molecular mechanisms involved in comorbidity.

To analyse the growing complexity of the graph, multiple methods were developed to reduce the graph under certain criteria. Any disease pair can be provided resulting in the subgraph connecting both diseases. This subgraph is often too large for manual analysis. A next step is the focus on another entity, like a drug or gene of interest, which again finds the connecting subgraph. The image below shows an example for asthma, diabetes, and Theophylline. Finally, a method was developed to automatically generate alle relevant subgraphs for all disease pairs and respective graph images for further analysis (Figure 1 [Fig. 1]).

Discussion: Discovering molecular factors in the pathogenesis of comorbid diseases is still complicated, but necessary for decision-making of treatment strategies. The addition of more diseases and sources poses challenges like uncertain or wrong data and mapping between entities and increases the analysis complexity. The new version of the GenCoNet pipeline is a vital step towards understanding comorbid diseases and their relationship to drugs and an ongoing process. First analyses showed shared genes and regulatory networks between asthma and diabetes. New relationships connecting previous findings could further inform the experimentally validated results from the cooperation. Finally, more sources and quality controls will be added in the future.

The authors declare that they have no competing interests.

The authors declare that an ethics committee vote is not required.