The use of dummy data points when fitting bacterial growth curves

L.A. Kelly, G. Gibson, G. Gettinby, W. Donachie, J.C. Low

Research output: Contribution to journalArticle

Abstract

We consider the problem of fitting mathematical models for bacterial growth and decline to experimental data. Using models which represent the phases of the growth and decline cycle in a piecewise manner, we describe how least-squares fitting can lead to potentially misleading parameter estimates. We show how these difficulties can be overcome by extending a data set to include hypothetical observations (dummy data points) which reflect biological beliefs, and the resulting stabilization of parameter estimates is analysed mathematically. The techniques are illustrated using real and simulated data sets.
LanguageEnglish
Pages155-170
Number of pages15
JournalIMA Journal of Mathematics Applied in Medicine and Biology
Volume16
Issue number2
DOIs
Publication statusPublished - Jun 1999

Fingerprint

Growth Curve
growth curve
microbial growth
least squares
mathematical models
Growth
Least-Squares Analysis
Least Square Fitting
stabilization
Theoretical Models
Stabilization
Mathematical models
Estimate
methodology
Experimental Data
Mathematical Model
parameter
Datasets
Model

Keywords

  • bacterial-growth modelling
  • mechanistic models
  • empirical models
  • least squares
  • dummy data points

Cite this

@article{5f7b7d4d2eda40abaf7a810e5bc54675,
title = "The use of dummy data points when fitting bacterial growth curves",
abstract = "We consider the problem of fitting mathematical models for bacterial growth and decline to experimental data. Using models which represent the phases of the growth and decline cycle in a piecewise manner, we describe how least-squares fitting can lead to potentially misleading parameter estimates. We show how these difficulties can be overcome by extending a data set to include hypothetical observations (dummy data points) which reflect biological beliefs, and the resulting stabilization of parameter estimates is analysed mathematically. The techniques are illustrated using real and simulated data sets.",
keywords = "bacterial-growth modelling, mechanistic models, empirical models, least squares, dummy data points",
author = "L.A. Kelly and G. Gibson and G. Gettinby and W. Donachie and J.C. Low",
year = "1999",
month = "6",
doi = "10.1093/imammb/16.2.155",
language = "English",
volume = "16",
pages = "155--170",
journal = "Mathematical Medicine and Biology",
issn = "1477-8599",
number = "2",

}

The use of dummy data points when fitting bacterial growth curves. / Kelly, L.A.; Gibson, G.; Gettinby, G.; Donachie, W.; Low, J.C.

In: IMA Journal of Mathematics Applied in Medicine and Biology, Vol. 16, No. 2, 06.1999, p. 155-170.

Research output: Contribution to journalArticle

TY - JOUR

T1 - The use of dummy data points when fitting bacterial growth curves

AU - Kelly, L.A.

AU - Gibson, G.

AU - Gettinby, G.

AU - Donachie, W.

AU - Low, J.C.

PY - 1999/6

Y1 - 1999/6

N2 - We consider the problem of fitting mathematical models for bacterial growth and decline to experimental data. Using models which represent the phases of the growth and decline cycle in a piecewise manner, we describe how least-squares fitting can lead to potentially misleading parameter estimates. We show how these difficulties can be overcome by extending a data set to include hypothetical observations (dummy data points) which reflect biological beliefs, and the resulting stabilization of parameter estimates is analysed mathematically. The techniques are illustrated using real and simulated data sets.

AB - We consider the problem of fitting mathematical models for bacterial growth and decline to experimental data. Using models which represent the phases of the growth and decline cycle in a piecewise manner, we describe how least-squares fitting can lead to potentially misleading parameter estimates. We show how these difficulties can be overcome by extending a data set to include hypothetical observations (dummy data points) which reflect biological beliefs, and the resulting stabilization of parameter estimates is analysed mathematically. The techniques are illustrated using real and simulated data sets.

KW - bacterial-growth modelling

KW - mechanistic models

KW - empirical models

KW - least squares

KW - dummy data points

UR - http://imammb.oxfordjournals.org/cgi/content/short/16/2/155?rss=1&ssource=mfc

UR - http://dx.doi.org/10.1093/imammb/16.2.155

U2 - 10.1093/imammb/16.2.155

DO - 10.1093/imammb/16.2.155

M3 - Article

VL - 16

SP - 155

EP - 170

JO - Mathematical Medicine and Biology

T2 - Mathematical Medicine and Biology

JF - Mathematical Medicine and Biology

SN - 1477-8599

IS - 2

ER -