Clinical bioinformatics for complex disorders: a schizophrenia case study

Abstract

Background: In the diagnosis of complex diseases such as neurological pathologies, a wealth of clinical and molecular information is often available to help the interpretation. Yet, the pieces of information are usually considered in isolation and rarely integrated due to the lack of a sound statistical framework. This lack of integration results in the loss of valuable information about how disease associated factors act synergistically to cause the complex phenotype.

Results: Here, we investigated complex psychiatric diseases as networks. The networks were used to integrate data originating from different profiling platforms. The weighted links in these networks capture the association between the analyzed factors and allow the quantification of their relevance for the pathology. The heterogeneity of the patient population was analyzed by clustering and graph theoretical procedures. We provided an estimate of the heterogeneity of the population of schizophrenia and detected a subgroup of patients featuring remarkable abnormalities in a network of serum primary fatty acid amides. We compared the stability of this molecular network in an extended dataset between schizophrenia and affective disorder patients and found more stable structures in the latter.

Conclusion: We quantified robust associations between analytes measured with different profiling platforms as networks. The methodology allows the quantitative evaluation of the complexity of the disease. The identified disease patterns can then be further investigated with regards to their diagnostic utility or help in the prediction of novel therapeutic targets. The applied framework is able to enhance the understanding of complex psychiatric diseases, and may give novel insights into drug development and personalized medicine approaches.

Figure 1

The complex disease network. Network representation of the information available in complex diseases. The top layer represents patient information and links between patients reflect associations between patients such as family relationships, sexual contact, ethical background or geographical proximity. The lower layer contains information about factors underlying the complex phenotype as quantified in genetic studies, molecular profiling experiments, clinical data etc. There is a strong dependency structure between the two layers that connect individual patients to abnormalities in the lower layer. The network concept thus allows the representation of available information in a patient specific manner.

Figure 2

Exploratory analysis – FANOVA. F-values and FDR adjusted p-values of the variables contained in clinical dataset. Several variables remain significant after controlling the FDR.

Figure 3

Disease network of a patient subgroup. The network reflects a remarkable abnormality in serum levels of primary fatty acid amides found in a subgroup of patients.

Figure 4

Degree distributions of patient and molecular networks. Bipartite modularity maximization guided by Markov Clustering algorithm determined a subgroup of patients (panel A) that hat highly increased node degrees (blue line) as compared to the remaining patients (red line). The associated molecules found in the same cluster (blue line, panel B) featured the same increased node degree. The node degree of the molecular analytes was found to follow a power law distribution (panel C).

Figure 5

Stability of Networks in schizophrenia and affective disorder. Comparison of the stability of the primary fatty acid networks determined from schizophrenia and affective disorder patients. The stability is measured using the networks' entropy during the clustering procedure. The entropy decreased at a lower rate in affective disorder, reflecting higher stability and a stronger alteration of the primary fatty acid network.

Figure 6

Mapping patients on the "diseasome" network. The spectrum of available data in complex diseases can be imagined as forming a "disease space" within which every disease occupies a particular location (left panel). In terms of networks, each disease forms a node and related diseases are connected by links reflecting the strength of association. The diagnostic process can be imagined as mapping a patient on the disease space or the "diseasome" network.