Algorithms in biology CS374 Fall 2006 Lecture 6, 10/12/06 Lecturer: Susan Tang Scribe: Neda Nategh Properties of Interaction Networks Based on the following papers: Annotation Transfer Between Genomes: Protein-Protein Interologs and Protein-DNA Regulogs Yu et al. Evidence for dynamically organized modularity in the yeast protein-protein interaction network Han et al. Introduction A. Motivation Protein interactions are ubiquitous and essential for cellular functions. For example, the processes such as signal transduction, metabolic pathway and transcription regulation are done by interaction of proteins. Here we explain briefly how interactions between proteins play the role in these processes. I. Cell signaling PIP2 cleavage to IP3 and DAG initiates intracellular calcium release and PKC activation One example for a cell signaling system is phosphatidylinositol (PIP2 ) metabolism and lipid signaling system. Phospholipase C is a key enzyme in this system that hydrolyzes PIP2, into two second messagers, inositol triphosphate (IP3) and diacylglycerol. During cellular signaling it regulates the activity of downstream proteins.Algorithms in biology CS374 Fall 2006 Lecture 6, 10/12/06 Lecturer: Susan Tang Scribe: Neda Nategh II. Metabolic Pathway There are several metabolic pathways; what we have chosen here, is Pre-mRNA Processing (Splicing). During RNA processing, snRNPs loop the intron and bind to it to form a spliceosome . Then the intron is separated from the exons and is removed from the pre-mRNA. Then exons are spliced together and produce the translatable mRNA. This mature mRNA exit the nucleus and is translated in the cytoplasm. III. Transcription Regulation The enzyme for transcription process needs some proteins associated with RNA polymerase, called transcription factor. The transcription factors bind to RNA polymerase or to another transcription factor or to cis-acting DNA sequences. A transcriptional regulatory network is a collection of these regulatory proteins associated with genes across a genome. A common transcription factor used for yeast two-hybrid screening is GAL4. In these examples we did not have that many interactions and we could use a tutorial diagram to visualize the interactions. But it is not possible to show all the interactions that occur within a cell in one diagram and the alternative way is to use the graph representation to convey all the interactions that occur within a cell. In graph representation, nodes are proteins and each of the edges represents an interaction.Algorithms in biology CS374 Fall 2006 Lecture 6, 10/12/06 Lecturer: Susan Tang Scribe: Neda Nategh There are two definitions for the interaction: 1. Physical Interactions, in which two proteins contact each other 2. Complex Interactions; it is the super set which means it contains all the physical Interactions; but it also contains the interactions in which two proteins interact but not directly. In some networks they show only physical interactions but they may show all the complex interactions in a network. B. Importance of Protein Interaction Networks Studying protein interaction network architecture allows us to: • Assess the role of individual proteins in the overall pathway: You may know the protein's function or how it is involved in a biological process, nevertheless you can not know the mechanisms by which the protein function until you have the entire network. By looking at the entire network you can realize which proteins are correlating. • Evaluate redundancy of network components: Suppose we have two proteins that communicate via a third protein; but they maybe can communicate without the third protein; but we can not know that utill we have an overal picture of the entire network. • Identify candidate genes involved in genetic diseases: If I have a disease and I know some genes involved in the disease but I do not have the entire set of associated genes to that disease; If one protein interacts with a high percentage of these genes, say %75, then we can say this protein is involved in that disease also. • Sets up the framework for mathematical models: If you know the entire interaction network a sort of sets of the frameworks you can write with differential equations simulating the individual modules whithin the network to see how regulation occurs within the network. For complex systems, the actual output may not be predictable by looking at only individual components. C. Protein Interaction Data Previously, experiments tested proteins one by one; but nowadays we have high-throughput experiments. Two famous high-throughput experiments are Yeast 2 Hybrid Screens and Co-IP. Yeast 2 Hybrid Screen (Cytotrap System)Algorithms in biology CS374 Fall 2006 Lecture 6, 10/12/06 Lecturer: Susan Tang Scribe: Neda Nategh The yeast two-hybrid method is used to study the pairwise protein-protein interactions. Gal4 protein is a transcriptional factor composed of two parts, a DNA-binding domain that interacts with promoter and an activating domain which interacts with polymerase. If two proteins interact physically, DNA-binding domain of GAL4 activates transcription. Detection of protein-protein interactions is based on the fact that what hSos requires to be recruited to the membrane. A yeast strain contains a temperature-sensitive mutation in the yeast Ras guanyl nucleotide exchange factor, cdc25-2. The target protein, is fused to hSos, and cDNA is fused to a membrane-localization signal. the cells only grow at 25°C but not at 37°C. When these hybrid proteins are co-expressed in a cdc25-2 yeast strain, allows growth at 37°C. If the two fusion proteins interact, hSos can recruit to the membrane. However, different sources of errors can affect the accuracy of these high-throughput experiments: False positives / False negatives Self-activators: proteins are naturally recruited to the cell membrane Promiscuous proteins Protein concentration differences Lack of benchmark D. Network cross-comparison Pairs of proteins have been binned according to their shortest path in networks generated from Y2H and Co-IP data. The false-color map indicates bins with more (red) or fewer (blue) interactions than expected by chance. Bins enriched for true positives, false positives and true noninteractors are indicated. In this figure,
View Full Document
Unlocking...