This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
This paper will present the review of the various techniques involved with modelling cancer along with discussing the computational methods involved and understanding the basic underlying principles with respect to modelling cancer.
In this review we also intend to describe the computational methods used to modelling cancer which involve the concepts of genomics, proteomics, metabolomics and cell signalling and network theories. The modelling cancer based on systems biology is been discussed based on three important types, which are biophysics, statistical modelling and modelling for clinical predictions.
We then discuss the recent studies and future directions which include the discovery of novel biomarkers, pathway analysis and annotation methods for cancer research and gaps to be addressed by systems biology.
Keywords: Systems biology, cancer modelling, computational methods, integrative biology, computational biology
Since the beginning of the human genome project, it has been pushing scientists towards an exceptional interpretation of biology, which we can call as systems approach. As the number of databases to store the scientific data are increasing discovery projects. The integration between the scientific discovery and scientific experimentation driven by hypothesis, forms a mandatory framework for systems biology (Ideker T et al., 2001).
Systems biology is a scientific field which uses the principles and expertise from an extensive range of fields such as engineering, mathematics, computer science and biological sciences to model the structure and dynamics of biological systems (Kherlopian A et al., 2008).
In this review we are specifically focussed on the various approaches for modelling cancers using systems biology approaches. Various approaches used to model cancer, their principles and applications are described in next sections.
2. Systems biology and cancer:
Computational technology and mathematical models are extensively used by systems biology to stimulate composite biological networks. The aim is not only to study a single component level, but rather study the principles, processes and mechanisms. The knowledge acquired from the simulation data can be used to design experiments and to design a refined accurate models with detailed description of physical and biological reality (Kherlopian A et al., 2008).
There are currently many promising candidates for oncogenic mutations that cause tumour development, but there is a huge gap of technologies and approaches to transmit this data into therapeutic use. Cancer cells have many noted properties such as complex signalling, mutations, signalling pathways, interaction between pathways. As a result of this properties we can observe various phenotypic changes in cancer cells.
As stated by Prasasya and co-workers in 2011, a strict interpretation of pathways is complicated by the extensive relation between signalling pathways. The analysis of the signalling pathways can be done using established systems biology tools. Here the experimental data is collected and computational approaches will be used and mechanistic accurate model can be obtained to analyse the pathways (Prasasya R et al., 2011; Fox EJ et al., 2009; Hornberg JJ et al., 2006; Aldridge BB et al., 2006).
Similarly systems biology can be used to study various factors causing cancers such as transcriptional factors, genetic factors, mutational factors, kinetics of the cells, physics of the cell etc. The knowledge from experimental data will be used as basis and an accurate models for the systems in study will be obtained.
Existing Biological Assumptions
Input Model development Model refinement Output
Fig: Systems Biology strategy, adapted and modified from R. Laubenbacher et al., 2009
3. COMPUTATIONAL METHODS:
Systems biology comprises of wide array of computational methods. With the advent of high throughput technologies to acquire genomic, proteomic and metabolomics data it is possible to model a wider range of biological processes.
It has evolved quickly in the past decade by advancements of High throughput genomic, Transcriptome, proteomic and metabolomics technologies. It has come to effect as a effective backbone for use of genomic data to predict, generate and test various hypothesis and to expose new biological insights in cancer research (Huang Y et al., 2012).
Genomics is a discipline, where rDNA and sequencing techniques are used along with the bioinformatics tools to study the whole genome. Genomics include Transcriptome, Omic mapping approaches or functional genomics and proteomic approach, proteome, interactome, phenome and localizome.
Fig: Genomics and Systems biology, adapted and modified from hui Ge et al., 2003
Obtaining data from genomics, integrating the data and modelling the biological processes together which form the systems biology approach using genomics (hui Ge et al., 2003).
One of the study conducted by Toyoshima M et al., 2011 describes the use of functional genomics in identifying potential targets for the cancer.
Proteomics deals with studies related to functional characterization of expressed proteins. It complements other approaches of functional genomics which also includes microarray-based expression profiles, phenotypic profiles at the cellular level, systematic genetics, and small-molecule-based arrays.
There has been tremendous advances in generating large scale data for protein-protein interaction, cellular composition, activity of the proteins, protein profiling in cancers.
Metabolomics, was introduced by Fiehn et al 2000; L.M. Raamsdonk 2001, it is a technique which involves a global detection of many small molecules metabolites and to identify metabolic preprograming. It can be defined as quantification and analysis of dynamic change in the metabolites which are under culture or certain biological conditions.
Metabolomics is a high throughput metabolite analysis, which has shown to have an impact in cancer diagnosis, cancer recurrence and also prognosis. It also finds its important role in detecting novel biomarkers and develop cancer therapeutics.
With the advances in diagnostics and therapy will further enable imminent growth of metabolomics, especially in the field of cancer research, where there is an ominous need for more sensitive and affordable cost effective diagnostic tool and fast need for developing therapies and also to identify novel biomarkers and for accurate predicting the response to the treatment (Nagrath D et al., 2011).
3.4 CELL SIGNALLING AND NETWORK THEORIES:
Numerous signalling molecules such as proteins, lipids and ions have been recognised and after many years of research, way of their communications with each other with the aid of transduction pathways have been revealed. Cell signalling consists of several sequential actions comprising covalent modifications e. g. phosphorylation, conscription, allosteric activation/inhibition, and binding of proteins (Alberts et al., 2002).
As many interactions were identified, it was clear that signalling does not occur independently by parallel linear pathways, but it includes large and complex network of interacting signalling pathways (Weng et al 1999). Identifying the structure and inherent properties or regularly arising regulatory motifs ('network motifs) is challenge and also may give a functional opinion of the union of signalling networks than a molecular view (Yeger-Lotemet al., 2004).
Till date, many cancer genes have been well studied and also the composition of numerous pathways in which their gene products function and various ways the pathways interact/communicate (Hanahan and Weinberg, 2000). This enables to draw a large 'road maps' of the information which are generally different in various cancer cells (Hornberg J et al 2006).
4. Cancer Modelling:
In cancer research, systems biology has three discrete threads:
4.1 Modelling biology
The goal of modelling biology or biophysics is to establish credibility or to obtain insights into the biologic process. Investigators will postulate a mathematical models along with parameters and run the model to check the qualitative mimics, as often there is no prescribed attempt to estimate the parametric data. This approach of systems biology is exploratory (Baker and Cramer, 2011).
This approach is seen to applied in various cancer research, such as detailed mathematical construct for cancer (Ao et al., 2008), investigation of cell assembly into tissue (Lemon et al., 2006), study of morphostats (Soto and Sonnenschein, 2004; Potter, 2007; Sonnenschein and Soto, 2008).
4.2 Statistical Modelling or Gene expression data
This approach shares the ambitious goal of systems biology to identify system mechanisms and connections from the gene expression data also referred to as 'reverse engineering' (Wang et al., 2007). This modelling systems include mathematical models which are based on molecular kinetics or statistical relations.
There are three main types of statistical relations:
Pair wise associations.
Graphical Gaussian models.
There are few challenges in regards of reverse engineering which use the gene expression microarray data:
Large spare networks.
Lack of statistical duplication.
Assumption made on biological systems (Baker and Cramer, 2011).
4.3 Modelling with the objective of clinical likelihoods
This strategy is used to recognise genetic networks from protein networks and assigning scores to this network and substitute this score to predict model, such as logistic regression (Chuang et al., 2007; Taylor et al., 2009). The predicted model will be formulated in training sample and will be evaluated in test sample.
This approach of modelling can be seen used by, Chuang et al., 2007, they have derived gene networks using protein networks which were used to predict metastatic and non-metastatic cancer in two studies. The have compared it for gene markers.
Taylor et al., 2009, used this approach to predict good and poor outcomes to predict cancer. They have predicated based on networks and clinical covariates. Although very few are used in clinical practice, it is still potential are for further research (Baker and Cramer, 2011).
5 Future Direction and Gaps to be addressed:
5.1 Systems Strategy for Marker Identification:
Markers are the molecular feature which can be used for diagnosis and prognosis of diseases with high accuracy e.g. DNA, RNA, proteins etc. unlike traditional gene based analysis , systems strategy is based on networks, identifying sub networks or modules such as markers.
Novel strategies and algorithms are been uncovered for cancer research, such as, pattern mining algorithm which detects cancer related functional sub network in human protein interactions (PPI) which occur in cancers (Shen et al., 2012).
Another network related gene ranking framework which is used to identify patient's genes of orphan disease or disease affecting small population. This algorithm is developed on Vertex similarity (VS) and parameter free (Zhu et al., 2012).
5.2 GWAS: Genome Wide Association Studies:
Since its first publication in 2005, GWAS have become an important area of research in systems biology. The major objective of GWAS is identification of genetic variants which are associated with physical trait or disease which is in investigation through statistical analyses of number of single nucleotide polymorphisms (SNP's) in a single experiment. Network and pathway based studies are major players in GWAS (Huang Y et al., 2012).
5.3 Gaps to be addressed:
The long term goal of genomic approaches for cancer research is to current diagnostic approaches and implement with mechanistic strategies of systems biology (Paules R et al., 2011). Biomarker studies are seen to have much importance in the field of systems biology as developing the biomarkers will be able to monitor exposure in humans which is required for risk assessment, various approaches for developing genomic biomarkers for patients exposed to specific agent will provide the required developments for developing wide array of biomarker based systems approaches (McHale et al., 2010).
Fig: Roadmap for systems biology for Cancer Research, adapted and Modified from Paules R et al., 2011
To obtain a detailed understanding of cancer, it is very necessary to utilize the expertise of various fields and develop an effective strategy to understand these events. As stated by D. Hanahan et al 2000, it will be possible to apply mathematical models to explain genetic function to reprogram the cell circuit to evident cancer. This review focusses on the advances and principles of systems biology approaches to modelling cancer, with the use of advances in mathematical and statistical modelling. Quantitative tools are also helping to provide insights into the integrated cell networks and signalling, and use of biomarkers in the form of novel diagnostic tool, modified or new treatments and also to develop novel drug targets for treatment of cancer (R. Laubenbacher et al., 2009).
The role of systems biology in cancer is inevitable to upturn with the association of various fields such as mathematics, biology, bioinformatics and high throughput techniques. And working together in interdisciplinary way, will inspire to advance in fundamental understanding.