SEARCH

SEARCH BY CITATION

Keywords:

  • systems biology;
  • cytomics;
  • cellular heterogeneity;
  • spatiotemporal models;
  • image databases

CONVERGENCE OF SYSTEMS BIOLOGY AND CYTOMICS

  1. Top of page
  2. CONVERGENCE OF SYSTEMS BIOLOGY AND CYTOMICS
  3. SPACE IN SYSTEMS BIOLOGY
  4. SHARING CYTOMICS DATA
  5. LITERATURE CITED

The current focus of systems biology is on the reverse engineering of networks to model gene regulatory or protein–protein interactions and the extraction of basic principles of biological organization. The ability of in silico representations to predict of how a system being in a particular state may react and adjust to perturbations has made systems biology an attractive component for basic research, drug development, and predictive medicine. However, computational systems biology is less experienced in implementing spatiotemporal properties of cells and multicellular architectures, and attempts to integrate and interconnect various levels of biological organization, such as genes, proteins, cells, and tissues, are a rare occurrence.

As opinions about the nature of Cytomics as a discipline arise, and activities and projects evolve, more specific definitions may become available. At this point it should be noted that even systems biology, which already has a track record of successful research activities, still experiences debates of what it is or should be. Cytomics is currently centered on high-throughput, multiparametric imaging in conjunction with machine vision to quantify cellular morphologies and properties like protein activities, with the underlying notion of a hypothesis-free approach. It aims to provide comprehensive, accurate, unbiased and systematic data, features that have been defined as the cornerstones for measurement technologies in systems biology (1). Applications for a systematic profiling of different cells, such as all cell types in the human body, have been suggested (2). Systematic screening projects will enable us to populate a rather sparse data area above the level of the proteome.

Cytomics appreciates single cell properties, which is of great value since averaging can seriously limit systems biology approaches, such as network analyses. The example recently demonstrated applied reverse engineering in T-cells, using multiparameter flow cytometry and Baysian statistics (3). Targeting multiple protein activators and inhibitors allowed simultaneous monitoring of the phosphorylation status of many protein species. As each pathway in each cell is in a particular status of activation, signaling networks can be more accurately revealed by taking multiparameter snapshots of many cells when compared to procedures that rely on averaged information out of cell lysates.

SPACE IN SYSTEMS BIOLOGY

  1. Top of page
  2. CONVERGENCE OF SYSTEMS BIOLOGY AND CYTOMICS
  3. SPACE IN SYSTEMS BIOLOGY
  4. SHARING CYTOMICS DATA
  5. LITERATURE CITED

Cytomics introduces something new that systems biology has not thoroughly considered yet: The ability to image makes quantitative spatial (and time resolved) measures of protein distributions and local concentrations, patterns related to organelles, and properties of transport phenomena between intracellular compartments available. Related projects include single cell-based response evaluation to drug treatment (4), pattern recognition of localized protein distributions that improve currently available ontologies (5), as well as analysis of changes of subcellular phenotypes due to systematic RNAi (6). In contrast, most currently performed modeling approaches in systems biology do not assume spatial distribution patterns and heterogeneities, but rather rely on a “well stirred” approximation. Pathway representations may reveal relationships between species but are currently disjoint from any spatiotemporal (4D) realities. Neglecting space and subcellular heterogeneities may limit our understanding of many cellular phenomena. As examples, computer representations of signaling networks such as the MAPK pathway, which take only diffusion into account, predict a rapidly dropping kinase concentration that would restrict signals reaching the nucleus. Therefore, realistic simulations require modeling of active transport processes, that can be measured experimentally (7). Molecular crowdings may play an important role, but need to be considered as unequally distributed in cellular compartments (8). Many cellular activities are based on localized protein concentrations and the concert of distinct activity patterns in cellular organelles (9), culminating in specific phenotypical responses.

As Cytomics in conjunction with systems biology finds some traction in analyzing cells and cell populations systematically, the next step envisions to extend these concepts for the investigation of relationship of cells in tissue constructs and in organ systems, both through comprehensive analysis and modeling. Some observations carried out in the past have taught us interesting lessons. Mina Bissell, LBNL, Berkeley, has noted that the structure of the tissue is dominant over the genome, and that a new paradigm for studying regulation of epithelial-specific genes in cancers is needed (10). Likewise, correlative multi-omics data analysis stratifying individual responses (11) and multiscale computer models that consider cellular heterogeneity in tissues will allow us to investigate metabolic diseases, immunity and aging, attempting to integrate available molecular, cellular, and physiological data (12).

SHARING CYTOMICS DATA

  1. Top of page
  2. CONVERGENCE OF SYSTEMS BIOLOGY AND CYTOMICS
  3. SPACE IN SYSTEMS BIOLOGY
  4. SHARING CYTOMICS DATA
  5. LITERATURE CITED

A particular challenge arising out of Cytomics is the need to store and share the information over the web, which in its core will be image-based. Microscopic databases so far have shown little acceptance; it's not the lack of technology, rather missing incentives or convincing arguments on their usefulness. Feasibility has long been demonstrated in pharma companies and radiology departments that entertain organization-wide image data warehouses. Cytomics will change this situation down the road, because more systematic, standardized, and collaborative projects will be defined that bring along scientific value to publicize, share, and standardize the representations resulting out of cell-based phenotyping projects. The location proteomics database currently under development by the Murphy group at Carnegie Mellon, Pittsburgh, is the first example of this kind (see http://murphylab.web.cmu.edu/services/PLSID). It is most likely that such image databases will be available soon for concurrent use with genomics and proteomics databases.

In conclusion, Cytomics has entered an interesting mutual relationship with systems biology that propels multiscale, systematic approaches. This venue will require close cooperation between experimentalists, imaging specialists, as well as modeling engineers and theoretical biologists to study biological complexity integratively.

LITERATURE CITED

  1. Top of page
  2. CONVERGENCE OF SYSTEMS BIOLOGY AND CYTOMICS
  3. SPACE IN SYSTEMS BIOLOGY
  4. SHARING CYTOMICS DATA
  5. LITERATURE CITED