Dynemo is a proprietary method of assessing changes in protein-interaction networks directly from gene expression data.
Gene expression
Cells in a living organism produce thousands of proteins for various purposes such as building and maintaining the structure of the cell, communicating within the cell and outside the cell, and as catalysts for metabolic and other processes. Cells produce proteins in a multi-step process ultimately using genes as the blueprint, i.e. proteins are the ultimate products of genes. Briefly, cells transcribe genetic DNA in the nucleus into RNA, which is transported to the protein manufacturing machinery of the cell and translated into proteins.
Studies that are aimed at understanding how proteins interact and how alterations in the quantities of proteins lead to disease typically use gene expression microarray technology. Microarrays allow for observation of thousands of gene products at once, providing a snapshot of what a cell is doing (which proteins are being produced and in what quantity). They are used to identify all the RNA that is transcribed in the nucleus at a point in time, and more importantly, the quantity of that RNA. Since most of the RNA is eventually translated into proteins the microarray provides a snapshot of which genes are active, and what proteins are acting in the cell.
Network biology
Proteins in a cell rarely act in isolation; each protein interacts with many other proteins to perform a specific function. Almost every protein in a cell is connected, via one more degrees of separation, creating a complex network of interactions. In a power grid, major disruptions in certain key nodes can have catastrophic effects and similarly, in a protein-protein interaction (PPI) network, disruptions in the organization of the network can lead to significant changes in the behaviour of the cell.
Dynemo
Using gene expression data, Dynemo is able to detect changes in the global organization of the PPI network by comparing the expression levels of interacting proteins. From observation of a of a gene expression data set taken from many patients, for example tumour samples and corresponding five-year survival data, Dynemo can reveal patterns in the tumour PPI networks that are associated with cancer severity. Once Dynemo’s algorithms have been trained to recognize these patterns in the sample set, they can be used to predict cancer severity in new patients. Beyond the example of cancer prognosis, Dynemo’s algorithms can be used to understand and predict other clinical outcomes, such as drug responsiveness. Since drugs target proteins, which interact in a network, the organization of the PPI network can alter the drug’s efficacy. Furthermore, Dynemo can facilitate the design and monitoring of combinatorial drug regimens that target multiple interacting proteins known to be disrupted in a particular disease state.
Gene expression
Cells in a living organism produce thousands of proteins for various purposes such as building and maintaining the structure of the cell, communicating within the cell and outside the cell, and as catalysts for metabolic and other processes. Cells produce proteins in a multi-step process ultimately using genes as the blueprint, i.e. proteins are the ultimate products of genes. Briefly, cells transcribe genetic DNA in the nucleus into RNA, which is transported to the protein manufacturing machinery of the cell and translated into proteins.
Studies that are aimed at understanding how proteins interact and how alterations in the quantities of proteins lead to disease typically use gene expression microarray technology. Microarrays allow for observation of thousands of gene products at once, providing a snapshot of what a cell is doing (which proteins are being produced and in what quantity). They are used to identify all the RNA that is transcribed in the nucleus at a point in time, and more importantly, the quantity of that RNA. Since most of the RNA is eventually translated into proteins the microarray provides a snapshot of which genes are active, and what proteins are acting in the cell.
Network biology
Proteins in a cell rarely act in isolation; each protein interacts with many other proteins to perform a specific function. Almost every protein in a cell is connected, via one more degrees of separation, creating a complex network of interactions. In a power grid, major disruptions in certain key nodes can have catastrophic effects and similarly, in a protein-protein interaction (PPI) network, disruptions in the organization of the network can lead to significant changes in the behaviour of the cell.
Dynemo
Using gene expression data, Dynemo is able to detect changes in the global organization of the PPI network by comparing the expression levels of interacting proteins. From observation of a of a gene expression data set taken from many patients, for example tumour samples and corresponding five-year survival data, Dynemo can reveal patterns in the tumour PPI networks that are associated with cancer severity. Once Dynemo’s algorithms have been trained to recognize these patterns in the sample set, they can be used to predict cancer severity in new patients. Beyond the example of cancer prognosis, Dynemo’s algorithms can be used to understand and predict other clinical outcomes, such as drug responsiveness. Since drugs target proteins, which interact in a network, the organization of the PPI network can alter the drug’s efficacy. Furthermore, Dynemo can facilitate the design and monitoring of combinatorial drug regimens that target multiple interacting proteins known to be disrupted in a particular disease state.