Integrated Cellular Pathology—Systems Biology of Human Diseases

We introduce a special issue focusing on articles presented at the meeting Cell Signal—Omics 2011—Integrated Cellular Pathology—Systems Biology of Human Disease, held in Luxembourg. The highly interesting and cutting-edge research results presented during this conference clearly demonstrate the power of large-scale experiments on complex biological systems, combined with high-throughput multiomics screening including genomics, transcriptomics, proteomics, metabolomics, or lipidomics. Indeed, Omics approaches have radically changed and improved biomedical research, and this conference attests to the power of data intensive biology (Hey et al., 2011).

The past 15 years brought hitherto unprecedented transformations in conceptualization of biological systems. Acceleration of communication and computing resources allowed an exponential production of results. Although the field of data intensive Omics science has initially focused only on genomics and transcriptomics, new interests and technical advances in high-throughput technologies have expanded the field that has now been enriched with proteome, miRome, Methylome, Surfaceome, Interactome, Degradome, and N-C terminome approaches. Systems became faster and more accurate allowing understanding complex biological systems. Today, new targets for the advancements of the Omics sciences include molecular state characterization (modification, stability, and location) and the new challenge of studying biology through integrating knowledge from the whole organism to the molecular state level.

Interesting talks related to the mathematical modeling of perturbations in cells were presented at this conference. Based on concepts originally from engineering and mathematics, a major focus is to conceptualize cellular phenotypes, including disease states, as stable network states. Accordingly, cells can be shifted from one state to another by activation of molecular switches, meaning that small changes can trigger cells to shift from their normal equilibrium state to a disease state equilibrium. Many molecular switches can coexist in cells, their signal propagation power being proportional to their connectivity. This way of conceptualizing cells highlights the multivariate state of cell regulation and can explain cell robustness or sensitivity to perturbations. For example, we have seen a very nice illustration by Antonio Del Sol (University of Luxembourg), explaining the conversion of benign forms of prion protein (PrPC) to disease-specific isoforms (PrPSc), to illustrate an example of a protein population shift, leading to a disease-related network perturbation.

In this quickly developing field, we have seen a very interesting presentation from Andrei Zinovyev (Institut Curie, INSERM U900, Paris, France) showing how to use the already available bibliographic information to build mathematical and logical reaction models, thus highlighting the interplay between several pathways. A global model of cell fate decision including survival, necrosis, and apoptotic pathways was shown, providing a comprehensive map of possibilities. These models can then be validated in silico or by new laboratory bench experiments challenging the key elements of the model and can be very helpful in suggesting new molecular targets to researchers.

Two presentations described high-throughput methods specifically developed to characterize posttranslational protein modifications. Although the cell membrane is a key interface of cell–environment interactions, membrane proteins are largely unknown, as they are difficult to characterize. In a first presentation, Bernd Wollscheid and colleagues (ETH, Zürich, Switzerland) developed a technology combining mass spectroscopy with cell surface capturing (CSC) technology in order to identify and quantify the cell surface glycoproteome. This technique allows characterization of low abundance membrane proteins that are typically a limiting factor. By identification of N-glycosides, it has also the potentiality of refining the cell membrane map, allowing for more precise and robust cellular identification and characterization. Several Omics methods (2D DIGE, mRNA expression patterns, high-throughput sequencing) provide a presence/absence registry but protein presence does not reflect protein functionality. In a second presentation, Chris Overall (University of British Columbia, Vancouver, Canada) presented a new technique based on a combination of proteomics and mass spectroscopy that is able to characterize C- and N-terminal degradation products in order to identify protein posttranslational modifications. Such a method leads to a degradomics repertoire that can be used to identify new protease substrates but also as a marker of protein functional state.

Single-cell analysis is the new and exciting frontier of Omics, as it can resolve cell subpopulations coexisting in heterogeneous tissues including blood. We have seen that single cell techniques are very useful to unravel cell changes occurring in cellular subfractions that are obscured by the much stronger signal generated by nonresponding cells when using classical population averaging techniques. For example, stem cells are suspected to play an important role in cancer resistance but represent only a small cellular subpopulation, currently very poorly characterized. However, this kind of analysis operates with very weak signals that will have to cope with a high signal-to-noise ratio associated to high amplification gains and with a dramatic increase of the data volume generated.

Besides interaction between cells and their environment or between proteins, and in addition to all the exciting new technical and modeling advances described above, several presentations emphasized the importance of another kind of crucial interaction: communication between scientists implicated in a given project. These presentations pointed that Omics should be preceded and followed by a scientific discussion, and that data do not represent results per se, but that results can only derive from specific and integrated data. Although this point is not new, we have seen that projects implicate teams of highly specialized scientists. A good understanding between all actors should take place from the beginning to the end. Experimental design must ensure that the most pertinent data type will be retrieved. The multiplicity of techniques developed during the last 2 decades now allows retrieving a large collection of data, and selecting the most pertinent ones. Result outputs must be formatted clearly, especially in the case of Omics results where figures must provide information while staying readable, as nicely shown by Andrei Zinovyev (Institut Curie, INSERM U900, Paris, France). Future meetings at the intersection of cellular signaling and Omics are presented below: