Display options
Cite
Zhu X, Welling M, Jin F, et al. Predicting Simulation Parameters of Biological Systems Using a Gaussian Process Model. Stat Anal Data Min. 2012;5(6):509-522doi: 10.1002/sam.11163.
Zhu, X., Welling, M., Jin, F., & Lowengrub, J. (2012). Predicting Simulation Parameters of Biological Systems Using a Gaussian Process Model. Statistical analysis and data mining, 5(6), 509-522. https://doi.org/10.1002/sam.11163
Zhu, Xiangxin, et al. "Predicting Simulation Parameters of Biological Systems Using a Gaussian Process Model." Statistical analysis and data mining vol. 5,6 (2012): 509-522. doi: https://doi.org/10.1002/sam.11163
Zhu X, Welling M, Jin F, Lowengrub J. Predicting Simulation Parameters of Biological Systems Using a Gaussian Process Model. Stat Anal Data Min. 2012 Dec;5(6):509-522. doi: 10.1002/sam.11163. PMID: 23482410; PMCID: PMC3589996.
Copy Download .nbib Format: NLM
Share it on

Stat Anal Data Min. 2012 Dec;5(6):509-522. doi: 10.1002/sam.11163.

Predicting Simulation Parameters of Biological Systems Using a Gaussian Process Model.

Statistical analysis and data mining

Xiangxin Zhu, Max Welling, Fang Jin, John Lowengrub

Affiliations

  1. Department of Computing Science, University of California Irvine, Irvine, USA.

PMID: 23482410 PMCID: PMC3589996 DOI: 10.1002/sam.11163

Abstract

Finding optimal parameters for simulating biological systems is usually a very difficult and expensive task in systems biology. Brute force searching is infeasible in practice because of the huge (often infinite) search space. In this article, we propose predicting the parameters efficiently by learning the relationship between system outputs and parameters using regression. However, the conventional parametric regression models suffer from two issues, thus are not applicable to this problem. First, restricting the regression function as a certain fixed type (e.g. linear, polynomial, etc.) introduces too strong assumptions that reduce the model flexibility. Second, conventional regression models fail to take into account the fact that a fixed parameter value may correspond to multiple different outputs due to the stochastic nature of most biological simulations, and the existence of a potentially large number of other factors that affect the simulation outputs. We propose a novel approach based on a Gaussian process model that addresses the two issues jointly. We apply our approach to a tumor vessel growth model and the feedback Wright-Fisher model. The experimental results show that our method can predict the parameter values of both of the two models with high accuracy.

Keywords: Gaussian process; biological simulation; regression

References

  1. Cancer Res. 2002 Sep 15;62(18):5381-5 - PubMed
  2. Bull Math Biol. 1998 Sep;60(5):857-99 - PubMed
  3. Genetics. 1931 Mar;16(2):97-159 - PubMed
  4. IEEE Trans Med Imaging. 2003 Sep;22(9):1163-71 - PubMed
  5. Annu Rev Biomed Eng. 2006;8:233-57 - PubMed
  6. J Cell Biol. 2005 May 23;169(4):681-91 - PubMed
  7. Math Med Biol. 2003 Jun;20(2):135-81 - PubMed
  8. Cancer Res. 1996 May 1;56(9):2185-90 - PubMed
  9. Int J Med Inform. 1999 Feb-Mar;53(2-3):239-52 - PubMed
  10. J Theor Biol. 2008 Aug 7;253(3):524-43 - PubMed
  11. Bull Math Biol. 2002 Jul;64(4):673-702 - PubMed
  12. J Theor Biol. 2010 Jun 21;264(4):1254-78 - PubMed
  13. Nonlinearity. 2010;23(1):R1-R9 - PubMed
  14. Nat Rev Cancer. 2010 Aug;10(8):587-93 - PubMed
  15. Bull Math Biol. 2004 Nov;66(6):1785-819 - PubMed

Publication Types

Grant support