Integrative biological analysis for neuropsychopharmacology

Abstract

Although advances in psychotherapy have been made in recent years, drug discovery for brain diseases such as schizophrenia and mood disorders has stagnated. The need for new biomarkers and validated therapeutic targets in the field of neuropsychopharmacology is widely unmet. The brain is the most complex part of human anatomy from the standpoint of number and types of cells, their interconnections, and circuitry. To better meet patient needs, improved methods to approach brain studies by understanding functional networks that interact with the genome are being developed. The integrated biological approaches--proteomics, transcriptomics, metabolomics, and glycomics--have a strong record in several areas of biomedicine, including neurochemistry and neuro-oncology. Published applications of an integrated approach to projects of neurological, psychiatric, and pharmacological natures are still few but show promise to provide deep biological knowledge derived from cells, animal models, and clinical materials. Future studies that yield insights based on integrated analyses promise to deliver new therapeutic targets and biomarkers for personalized medicine.

Figure 1

Computational analysis of comprehensive proteomic, transcriptomic, and lipidomic data sets can yield new insights derived from correlations hidden in the data.

Figure 2

Ultrahigh-resolution isotopic fine structure of a metabolite, S-methyl-l-cysteine. Spectra were collected with a 12 T FT-ICR MS with a mass resolving power of ∼340 K. The exact mass of S-methyl-l-cysteine [M+H]+is 136.04267. The zoomed inset shows the first ¹³C isotopic fine structure. The natural isotope abundances are: ¹⁵N, 0.36% ³³S, 0.76% ¹³C, 1.1% and ²H, 0.015%. The exact mass, isotopic fine structure, and isotope abundances are used to assign unambiguously elemental compositions.

Figure 3

The raw lipidomic data is a matrix of samples over variables. (a) The samples are the individual perturbations, which are grouped into control samples and the sample(s) of interest to the study, here simply called ‘disease'. Partial correlations of all variables are obtained and later evaluated with respect to statistical significance. (b) To investigate whether a significant partial correlation is specific for the disease sample, partial correlations (as in a) were calculated for the entire data set, as well as for data sets where each one sample was left out. Unless a correlation is significant in all Gaussian Graph Models (GGMs), it is considered disease-specific. This figure is reproduced with the permission of the publisher (Mueller et al, 2011).

Figure 4

Lipids specifically regulated when a glioma cell was effectively perturbed. (a) Relative numbers of disease-specific and -unspecific lipid–lipid partial correlations in the Gaussian Graph Models (GGMs). Analysis of the entire data set is named ‘conventional' GGM with respect to disease specificity. (b) Disease-relevant GGM, which is associated with the combined perturbation by WT-p53 adenoviral transfection before SN38 chemotherapy in U87 glioma cell lines. (c) Modularity matrix was calculated by using the lipid species as class label for the GGM in b. PE, phosphatidylethanolamine; PG, phosphatidylglycerol; PI, phosphoinositol; PS, phosphoserine. This figure is reproduced with the permission of the publisher (Mueller et al, 2011).

Figure 5

The clustered glycolipids in treatment with WP1193 (threshold of 0:9). The edges are colored on a linear HSV-scale (from blue=cold to red=hot) by the absolute value of the edge weight, and the size of the nodes is proportional to sum of their incident absolute edge weights. In this difference graph, a node is large if it differs strongly in its correlations to other nodes in the WP1193 treatment. Small nodes roughly preserve their correlation behavior in this treatment. This figure is reproduced with the permission of the publisher (Goerke et al, 2010).