lecture_20

lecture_20 - 6.874/6.807/7.90 Computational functional...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
1 6.874/6.807/7.90 Computational functional genomics, lecture 20 (Jaakkola) Models of transcriptional regulation We have already discussed four simple mechanisms of transcriptional regulation, nuclear exclusion nuclear concentration modiFcation of bound activator redirection of binding sites due co-regulator Much of transcriptional regulation depends on protein-protein interactions that are mod- ulated by protein modiFcations such as phosphorylation (we will only consider phospho- rylation). A single protein may have multiple phosphorylation sites and its activity may depend combinatorially on the state of the individual sites. To represent a protein state we introduce variables x P ( x = 0) 0 inactive P ( x = 1) 1 active where the two states (0/1) are tied to protein activity (whether it is phosphorylated). One way to build mechanisms is to use these variables as nodes in a Bayesian network. In many cases, however, Bayesian networks would not able to represent the mechanisms explicitly but would require appropriate setting of the parameters to characterize the be- havior of the interacting variables. So while Bayesian networks could effectively capture the mechanism, they would not yield an appropriate visual representation of the mechanism (the parameters are not visible in the graph). Moreover, several mechanisms may be con- sistent with a single graph structure while differing substantially in terms of the choice of the parameter values. We focus here instead on a complementary way of capturing protein states and their dynamics using state transition diagrams.
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
2 6.874/6.807/7.90 Computational functional genomics, lecture 20 (Jaakkola) Transition diagrams Protein states are often transient so our representation of state must involve time in some manner. We can, for example, explicate transitions from one state to another within some reference time interval: 1 0 inactive active where, in terms of phosphorylation, the transition from inactive to active state is due to kinase(s) while the reverse transition is facilitated by phosphatases. It is no longer sufficient to characterize the model in terms of static probabilities P ( x = 0) and P ( x = 1). Instead, we must consider the state
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 11/11/2011 for the course BIO 7.344 taught by Professor Bobsauer during the Spring '08 term at MIT.

Page1 / 8

lecture_20 - 6.874/6.807/7.90 Computational functional...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online