Dear Avi,
elaborating further also based on Jim's answer, some open questions/considerations that you might entertain are:


- if you intend to use sliding semilandmarks (which in general sounds like a good idea), how are certain points such as 1,39,41,55 going to be slid? (i.e., relative to which other points?) - Notice that in the paper you cited the points on the tail are bounded by fixed landmarks. In that respect, I wonder if maybe your point 1 could be characterized as a fixed homologous point (you define it as "dorsal connection of the tail fin")

- I guess that whether the relative position of the eye is useful depends on your particular biological question; as Jim said, it might actually be interesting; an alternative to using a fixed landmark in the eye (at the center of it, I assume) would be to use sliding semilandmarks along the contour of the eye. These can be slid relative to each other and can give you information on the relative size of the eye (which may or may not be of interest for your specific question). This seems the solution adopted by Levis et al. (at least looking at their figure). Another option with respect to that would also be removing the extra semilandmarks after sliding considering them "helper points" (see, for instance, fig. 1a in Fruciano et al. 2016 - Ecology and Evolution)

- If you can model variation due to arching with a relatively simple model, there are methods to account for it, including the one implemented in tpsUtil and the one developed by Valentin et al. 2008 - Journal of Fish Biology (you can find a discussion of these also in my review on measurement error Fruciano 2016 - Development Genes and Evolution). Collecting data as accurately as possible in the first place is the best idea (and you're doing that), but substantial artifactual variation might be still present in your data despite of your best efforts (if at all possible, consider also quantifying how similar repeated pictures of the same specimens are; you can read about this also in the review I mentioned above)

I hope this helps.
Best,
Carmelo


Il 28/02/2018 9:30 PM, Avi Koplovich ha scritto:
Hi James,
Thank you for your fast answer.
I'll continue to mark the landmarks as separated sublets, i.e. head and tail.

 1. Is it ok to use landmark 40 (intersection of the side line and the
    dorsal connection of the tail fin) as a common mark point to create
    a comb fan for both tail and head?
 2. Another question: Is this landmark 40 can be treated as fixed
    landmark as it is restricted by both x and y axis?
 3. Can I use landmarks 20 (tail tip) and 48 (head tip) as fixed
    landmarks in an ontogenetic experiment?
 4. Last question: Can/Should I use the eye as a fixed landmark for the
    head (i.e. can it interfere with interpreting the head contour)?

p.s. I saw your forum message here on updating tpsdig 2.31 and already updated it.


Many thanks,

Avi



On Wednesday, February 28, 2018 at 8:51:39 PM UTC+2, f.james.rohlf wrote:

    Bending of long slender organisms (or other structures) can be an
    important practical problem. Changes due to bending could dominate
    the results. Separating sets of landmarks into more rigid sublets
    can help as long as the endpoints of the subsets are homologous and
    not arbitrary or dependent on the bending of a particular specimen.
    The tpsUtil program has an "unending" option if one has a subset of
    points that one knows should be in a straight line. That would be
    difficult for larvae.

    Note: I see you are using ver. 2.30 of tpsDig. Version 2.31 is
    current. I do try to keep fixing bugs so It can be useful to stay
    current.

    ____________________________________________
    F. James Rohlf, Distinguished Professor, Emeritus. Ecology & Evolution
    Research Professor, Anthropology
    Stony Brook University

    -----Original Message-----
    From: Avi Koplovich [mailto:netbird....@gmail.com <javascript:>]
    Sent: Tuesday, February 27, 2018 3:09 AM
    To: MORPHMET <morp...@morphometrics.org <javascript:>>
    Subject: [MORPHMET] Digitizing landmarks on live larvae

    Hi,
    I've started a new project and came to the point of marking fixed
    and semi landmarks.
    Not all pictures are satisfying, mostly because of the posture of
    the larvae during photographing (sometimes raising it's tail). So in
    order to reduce the noise by the animal posture, I thought it would
    be helpful to separate head and tail as was done in Levis et. al.
    2016, Biol. J. Linn. Soc.
    I'm using the landmarks 1, 20 and 48 as fixed landmarks, and all the
    rest are semi landmarks. I'm not sure of using 20 and 48 as fixed
    landmarks, and I wonder if I can use landmark 40 as fixed landmark
    since it is restricted by both x (side line) and y (dorsal
    connection of the tail fin). Can/Should I use the eye as a fixed
    landmark for the head (i.e. can it interfere with interpreting the
    head contour)?
    Here is an example to show what I mean:
    
https://drive.google.com/file/d/1iO7lCN3ZCtV7DF9vsczkb_EYoSli1Orr/view?usp=sharing
    
<https://drive.google.com/file/d/1iO7lCN3ZCtV7DF9vsczkb_EYoSli1Orr/view?usp=sharing>

    I'll be happy if you can advise on that.
    Thank you,
    Avi

-- MORPHMET may be accessed via its webpage at
    http://www.morphometrics.org
    ---
    You received this message because you are subscribed to the Google
    Groups "MORPHMET" group.
    To unsubscribe from this group and stop receiving emails from it,
    send an email to morphmet+u...@morphometrics.org <javascript:>.

--
MORPHMET may be accessed via its webpage at http://www.morphometrics.org
---
You received this message because you are subscribed to the Google Groups "MORPHMET" group. To unsubscribe from this group and stop receiving emails from it, send an email to morphmet+unsubscr...@morphometrics.org <mailto:morphmet+unsubscr...@morphometrics.org>.

--
MORPHMET may be accessed via its webpage at http://www.morphometrics.org
--- You received this message because you are subscribed to the Google Groups "MORPHMET" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to morphmet+unsubscr...@morphometrics.org.

Reply via email to