Abstract
Certain biological experiments investigating cell motion result in time lapse video microscopy data which may be modeled using stochastic differential equations. These models suggest statistics for quantifying experimental results and testing relevant hypotheses, and carry implications for the qualitative behavior of cells and for underlying biophysical mechanisms. Directional cell motion in response to a stimulus, termed taxis, has previously been modeled at a phenomenological level using the Keller-Segel diffusion equation. The Keller-Segel model cannot distinguish certain modes of taxis, and this motivates the introduction of a richer class of models which is nevertheless still amenable to statistical analysis. A state space model formulation is used to link models proposed for cell velocity to observed data. Sequential Monte Carlo methods enable parameter estimation via maximum likelihood for a range of applicable models. One particular experimental situation, involving the effect of an electric field on cell behavior, is considered in detail. In this case, an Ornstein- Uhlenbeck model for cell velocity is found to compare favorably with a nonlinear diffusion model.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 23-37 |
| Number of pages | 15 |
| Journal | Journal of Mathematical Biology |
| Volume | 48 |
| Issue number | 1 |
| DOIs | |
| State | Published - Jan 2004 |
Fingerprint
Keywords
- Cell migration
- Chemotaxis
- Galvanotaxis
- Nonlinear diffusion
- Stochastic model
ASJC Scopus subject areas
- Agricultural and Biological Sciences (miscellaneous)
- Mathematics (miscellaneous)
Cite this
Stochastic models for cell motion and taxis. / Ionides, Edward L.; Fang, Kathy S.; Isseroff, Roslyn Rivkah; Oster, George F.
In: Journal of Mathematical Biology, Vol. 48, No. 1, 01.2004, p. 23-37.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Stochastic models for cell motion and taxis
AU - Ionides, Edward L.
AU - Fang, Kathy S.
AU - Isseroff, Roslyn Rivkah
AU - Oster, George F.
PY - 2004/1
Y1 - 2004/1
N2 - Certain biological experiments investigating cell motion result in time lapse video microscopy data which may be modeled using stochastic differential equations. These models suggest statistics for quantifying experimental results and testing relevant hypotheses, and carry implications for the qualitative behavior of cells and for underlying biophysical mechanisms. Directional cell motion in response to a stimulus, termed taxis, has previously been modeled at a phenomenological level using the Keller-Segel diffusion equation. The Keller-Segel model cannot distinguish certain modes of taxis, and this motivates the introduction of a richer class of models which is nevertheless still amenable to statistical analysis. A state space model formulation is used to link models proposed for cell velocity to observed data. Sequential Monte Carlo methods enable parameter estimation via maximum likelihood for a range of applicable models. One particular experimental situation, involving the effect of an electric field on cell behavior, is considered in detail. In this case, an Ornstein- Uhlenbeck model for cell velocity is found to compare favorably with a nonlinear diffusion model.
AB - Certain biological experiments investigating cell motion result in time lapse video microscopy data which may be modeled using stochastic differential equations. These models suggest statistics for quantifying experimental results and testing relevant hypotheses, and carry implications for the qualitative behavior of cells and for underlying biophysical mechanisms. Directional cell motion in response to a stimulus, termed taxis, has previously been modeled at a phenomenological level using the Keller-Segel diffusion equation. The Keller-Segel model cannot distinguish certain modes of taxis, and this motivates the introduction of a richer class of models which is nevertheless still amenable to statistical analysis. A state space model formulation is used to link models proposed for cell velocity to observed data. Sequential Monte Carlo methods enable parameter estimation via maximum likelihood for a range of applicable models. One particular experimental situation, involving the effect of an electric field on cell behavior, is considered in detail. In this case, an Ornstein- Uhlenbeck model for cell velocity is found to compare favorably with a nonlinear diffusion model.
KW - Cell migration
KW - Chemotaxis
KW - Galvanotaxis
KW - Nonlinear diffusion
KW - Stochastic model
UR - http://www.scopus.com/inward/record.url?scp=6044249273&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=6044249273&partnerID=8YFLogxK
U2 - 10.1007/s00285-003-0220-z
DO - 10.1007/s00285-003-0220-z
M3 - Article
C2 - 14685770
AN - SCOPUS:6044249273
VL - 48
SP - 23
EP - 37
JO - Journal of Mathematical Biology
JF - Journal of Mathematical Biology
SN - 0303-6812
IS - 1
ER -