文档介绍：Beyond Co-expression: work Inference
Patrik D’haeseleer
Harvard University
http://~patrik
Beyond Co-expression
Clustering approaches rely on co-expression of genes under different conditions
Assumes co-expression is caused by co-regulation
We would like to do better than that:
Causal inference
What is regulating what?
work Inference
Overview
Modeling Issues:
Level of biochemical detail
Boolean or continuous?
Deterministic or stochastic?
Spatial or non-spatial?
Data Requirements
Linear Models
Nonlinear models
Conclusions
Level of Biochemical Detail
Detailed models require lots of data!
Highly detailed biochemical models are only feasible for very small systems which are extensively studied
Example: Arkin et al. (1998), ics 149(4):1633-48
lysis-lysogeny switch in Lambda:
5 genes, 67 parameters based on 50 years of research, stochastic simulation required puter
Example: Lysis-Lysogeny
Arkin et al. (1998), ics 149(4):1633-48
Level of Biochemical Detail
In-depth biochemical simulation of . a whole cell is infeasible (so far)
Less work models are useful when data is scarce and/work structure is unknown
work structure has been determined, we can refine the model
Boolean or Continuous?
works (Kauffman (1993), The Origins of Order) assumes ON/OFF gene states.
Allows analysis at work-level
Provides useful insights work dynamics
Algorithms work inference from binary data
A
B
C
C = A AND B
0
1
0
Boolean or Continuous?
Boolean abstraction is poor fit to real data
Cannot model important concepts:
amplification of a signal
subtraction and addition of pensating for smoothly varying environmental parameter (. temperature, nutrients)
varying dynamical behavior (. cell cycle period)
Feedback control:
negative feedback is used to stabilize expression
causes oscillation in Boolean model
Deterministic or Stochastic?
Use of concentrations assumes individual molecules can be ignored
Known examples (in prokaryotes) where stochastic