AVS 59th Annual International Symposium and Exhibition
    Biofilms and Biofouling: Marine Medical Energy Focus Topic Thursday Sessions
       Session MB+BI-ThM

Paper MB+BI-ThM6
Analysis of Force Curves of Pseudomonas Aeruginosa obtained by Atomic Force Microscopy

Thursday, November 1, 2012, 9:40 am, Room 23

Session: Biofilms and Biofouling in Medicine
Presenter: N.A. Burnham, Worcester Polytechnic Institute
Authors: E.V. Anderson, Worcester Polytechnic Institute
R.L. Gaddis, Worcester Polytechnic Institute
T.A. Camesano, Worcester Polytechnic Institute
N.A. Burnham, Worcester Polytechnic Institute
Correspondent: Click to Email

Pseudomonas aeruginosa is extremely harmful to immune-compromised individuals. An atomic force microscope (AFM) can be used to measure the forces between the AFM tip and the bacterial exopolymers, with which the bacteria attach themselves to surfaces. These forces are characterized with a model that is a function of brush (exopolymer) layer thickness, probe radius, temperature, separation distance, and a molecular volume. Initial experiments with limited data sets are consistent with expected brush thicknesses of a few hundred nanometers. In order to progress – now with rigor – we have just developed a high throughput method for the analysis of force curves on the exopolymers of P. aeruginosa [1]. The above-described model is only valid for the region where the tip is in contact with the exopolymers, yet is not perturbing the bacterial membrane. MatLab code was written to determine the location of this region in each force curve, crop the curve to that region, and apply the force model in order to obtain parameters of the exopolymers. The standard deviation of the mean and Chauvenet’s Criterion are then applied to the results of sets of one-hundred force curves to increase measurement precision and objectively remove outliers. This procedure removes user subjectivity in cropping, fitting, and outlier removal, decreases analysis time by two orders of magnitude, and increases the precision of fitted results by a factor of ten (for one-hundred curves), which is necessary for demonstrating the statistical significance of our data.

[1] Anderson et al., to be submitted May 2012.