Hi,

I'm completely new to this field.  I'm wanting to look at variation in
optic disc shape (assess heritability) as a possible predictor for optic
nerve diseases (eg glaucoma).

Essentially, the basic form of an optic disc is a vertically shaped oval
(although some peoples discs are more circular and even (rarely)
horizontally oval).  The curves which make up the shape are not
completely smooth however, and usually there are subtle changes in
curvature around this 'oval' shape.  Also, optic discs usually are not
perfectly symmetrical.

I'm unsure how to tackle this shape problem.  Initially, I thought of
treating each disc as a smooth and symmetrical oval and calculating 2
curvatures (for the steep and flat curves of the oval (assuming
symmetry)) and describing this as some sort of shape or form factor.

Then I stumbled onto morphometrics.  Which leads me to ask 2 questions:

1. Would using geometric morphometrics be considered overkill (by those
who know enough about it) for this type of problem?  In my readings it
seems (Elliptic?) Fourier analysis could be used in this case.

2. Is there any way of converting this information into a meaningful
biological variable?  Can the numbers you get from whatever analysis you
do be used in a biological sense, or are they only meaningful when
describing the shape mathematically? (sorry if I haven't explained this
well).  Ultimately, I'm wanting to do genetic analyses on a sample of
people and want to be able to make associations with shape.

Thanks for your time,

Regards,

Paul




-- 
Replies will be sent to the list.
For more information visit http://www.morphometrics.org

Reply via email to