June 2022
Volume 63, Issue 7
Open Access
ARVO Annual Meeting Abstract  |   June 2022
Comparing methods for quantifying myosin distributions in the avian crystalline lens
Author Affiliations & Notes
  • Adeline Thea Suko
    School of Optometry and Vision Science, University of Waterloo, Waterloo, Ontario, Canada
  • Alexander Wong
    Systems Design Engineering, University of Waterloo, Waterloo, Ontario, Canada
    Waterloo Artificial Intelligence Institute, University of Waterloo, Waterloo, Ontario, Canada
  • Vivian Choh
    School of Optometry and Vision Science, University of Waterloo, Waterloo, Ontario, Canada
  • Footnotes
    Commercial Relationships   Adeline Suko None; Alexander Wong None; Vivian Choh None
  • Footnotes
    Support  NSERC Discovery Grant, Canadian Optometric Education Trust Fund
Investigative Ophthalmology & Visual Science June 2022, Vol.63, 656 – F0011. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Adeline Thea Suko, Alexander Wong, Vivian Choh; Comparing methods for quantifying myosin distributions in the avian crystalline lens. Invest. Ophthalmol. Vis. Sci. 2022;63(7):656 – F0011.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose : Disruption of the actomyosin networks in the chicken lens has been shown to affect lenticular biomechanics. This study compares methods of analyzing and quantifying myosin distributions.

Methods : Lenses from 7-day old white leghorn (Gallus gallus domesticus) chickens were labelled for myosin. Nearest neighbour distances (NNDs) for myosin centroids at the posterior surface were determined using three methods on the same image: 1) ImageJ: manual using the Elliptical tool, 2) ImageJ: automatic using the Analyze Particle tool, 3) MATLAB: automatic using custom-coded software. For analyses using ImageJ, the NND plugin was used. First, the Elliptical tool (manual) on ImageJ was used to determine XY centres of ellipses superimposed on the image by the user. Three ellipses were superimposed on each centroid to account for user error. For the automated ImageJ method, the image was converted to a binary format using the program's Make Binary function. The Fill Holes and Eraser tools in ImageJ were used to correct centroids altered by the thresholding process before determining the myosin centres using the Analyze Particles tool. For analysis in MATLAB, the image was opened and the custom-coded script was run to obtain the myosin centres and NNDs. Corrections were made using MATLAB's Brush tool. The three methods of calculating NNDs from the same 135 centroids were compared (repeated measures ANOVA).

Results : The Analyze Particles tool detected the fewest centroids (135) compared to the Elliptical (166) and MATLAB methods (163). Although the elliptical tool allowed for greater detection, the ellipses do not always fit properly, possibly leading to inaccurate NND values. The MATLAB program uses adaptive thresholding to consider variations in illumination, allowing it to detect more centroids with fewer detection errors than the Analyze Particles tool. Ranges and mean NNDs (±SD) for the same 135 centroids are listed in Table 1; no significant differences were found between the three methods (P=0.2754).

Conclusions : The custom MATLAB program is the choice method for mapping out the proten network using confocal images. It is automated, reducing user error compared to the manual elliptical tool. It also recognized more centroids than the ImageJ "Analyze Particles" feature with fewer errors. Future work involves actin detection to measure actin-myosin distances.

This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.

 

Table 1: Ranges and means for the same 135 NNDs.

Table 1: Ranges and means for the same 135 NNDs.

×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×