BIOL0029 2022 Homework 2: The evolution of skeletal anatomy and locomotor mode in frogs
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
BIOL0029 2022 Homework 2: The evolution of skeletal anatomy and locomotor mode in frogs
Background: Frog Locomotion
Frogs have anatomical specialisations of the limbs and pelvis, enabling them to employ a wide range of
locomotor modes, including terrestrial and arboreal jumping, swimming, walking, hopping, and burrowing
(Figure 1). Different frog species inhabit diverse environments, making them ideal subjects for investigating the relationship between anatomy, ecology, and evolutionary history (Jorgensen & Reilly, 2013).
Terrestrial jumper
(TJ)
Burrower-walker-hopper
(BWH)
Figure 1: Five different types of frog locomotor mode with representative species: Rana temporaria (TJ), Litoria chloris (AJ), Xenopus laevis (AQ), Brachycephalus ephippium (WH), Rhinophrynus dorsalis (BWH).
Locomotor performance has been linked to distinct pelvic features, the most prominent being the shape of
the sacral diapophyses (Figure 2; Emerson, 1979; Jorgensen & Reilly, 2013) . There are three pelvic types in
frogs, that characterise specific movements: lateral-bender (LB), sagittal-hinge (SH) and fore-aft slider (FA)
(Figure 2) . However, it is challenging to classify frogs into these groups, in particular distinguishing between
SH and LB (Soliz et al . , 2017) . Instead, pelvic anatomy may be better characterised as a set of continuous
multidimensional variables, rather than as three discrete types, suggesting more complex links between anatomy and function in frog skeletal structures than previously thought. One way to define this continuous
variation is by the degree of expansion of the sacral diapophyses (ESD; Figure 2). ESD is potentially the strongest driver of anatomical variation and, therefore, predictor of locomotor mode (Jorgensen & Reilly, 2013).
Another strong predictor of locomotor performance is the length of the hindlimbs (James & Wilson, 2008).
Longer hindlimbs are associated with jumping and swimming frogs, while frogs that walk, hop, and/or burrow
have shorter hindlimbs . However, the extent to which the length of each individual bone within the hindlimb
impacts locomotion is unclear (Enriquez-Urzelai et al., 2015; Lires et al., 2016. Each bone is hypothesized
Lateral-bender (LB)
Sagittal-hinge (SH)
Fore-aft slider (FA)
Expansion of
(ESD)
Figure 2: Three different types of frog pelvis (Emerson, 1979). ESD (blue arrow) is the distance between
the most anterior and posterior tip of the sacral bone (outlined in red).
to play different roles, as different hindlimb proportions will have different effects on hindlimb biomechanics.
Additionally, evolutionary history will have played a role in determining anatomy and locomotor mode.
The Assignment
The primary aim of this homework is to investigate the question: to what extent do locomotor and phyloge-
netic demands impact skeletal anatomy in frogs? Additionally, you will gain experience in a) independently
analysing a real dataset from beginning to end, and b) writing up your statistical analyses and their biological interpretation. Your report should be shaped by this question (ie not simply a description of analyses).
The dataset “BIOL0029.HW2.2022.txt” contains data from 164 species of frogs from across a broad range
of positions in the frog phylogeny. Each frog has six skeletal measurements (Figure 3) and two categorical variables – their locomotor mode (Figure 1) and their phylogenetic clade (the oldest group of extant frogs
(Basal) or a group that has evolved more recently (Neobatrachia, Hyloidea or Ranoidea)). The Homework breaks down into the following tasks.
1. To make meaningful comparisons between species, you should first control for body size by dividing each skeletal measurement by snout-vent length (SVL). Examine the resulting scaled measurements to ensure there are no outliers.
2. For each skeletal measurement, determine which locomotor modes and phylogenetic clades are sig- nificantly associated with it, using a p-value threshold of 0.05. Your analysis should consider the possibility of interactions between locomotor mode and clade. All the models should be fitted and compared using the R functions lm(), summary(),and anova(). Report which duos/trios of clade, loco- motor mode and measurement are associated (i.e., [clade X] is associated with both a larger [skeletal measurement 1] and [locomotor mode A], and a larger [skeletal measurement 1] and [locomotor mode A] are also associated with each other). Present these results imaginatively, using relevant tables or figures – don’t simply show the R output. Discuss what your findings suggest about the role of each bone in locomotion, referring to previous literature.
3. Perform a principal components analysis (PCA) of the scaled skeletal measurements, excluding SVL. Examine the loadings of the PCA model and discuss which measurement variables are the strongest drivers of morphological variation. Discuss how much variation is explained by each principal compo- nent.
SVL
Figure 3: Frog skeleton labelled with the anatomical variables measured in the dataset provided: snout-vent
length (SVL), expansion of the sacral diapophyses (ESD), femur (fem), tibiofibular (tib), calcaneus (calc) and
foot.
4. Plot your PCA, colour-coded for locomotor mode. Repeat for phylogenetic clade. Consider the use of biplots. Consider how much anatomical variation is in each group, and what that implies in terms of conservation across the phylogeny. Using linear models, investigate if there is a statistically significant relationship between each PC and locomotor mode and phylogenetic clade.
5. Summarise your results in the most informative and creative way possible. Use informative Tables and Figures. Based on your results and further reading, consider whether you can predict locomotor function or phylogenetic position using skeletal morphology alone.
The Report
Your report should be an R markdown document that contains three sections: ‘Introduction’ should be no
more than a short paragraph setting out the aims of the report and summarising the data . ‘Methods and
Results’ provides information on the statistical methods you used and the results you obtained, including
tables and figures. You should include a clear but brief description of the data. In the ‘Discussion’ you provide the biological interpretation (but no new results nor, typically, figures).
Your markdown should only include a final set of commands, not the code of any exploratory analyses that
you may have done along the way. It should run through all of the code (from loading the data through generating all your results and producing table and figure output) on compilation (i.e., when pressing ‘Knit’
in RStudio) . The code should assume that the data file is located in the same directory (folder) as the
markdown.
While containing all code, the file should be written such that the compiled PDF of your markdown contains
methods and results in a well-digested form. For example, you would not render visible any commands that are not essential for the understanding of your work (e.g., commands to load data or those generating the plots displayed). Also try to make the output that you do show (e.g., ANOVA tables or estimated coefficients) as pretty as possible, by cutting out unnecessary output normally produced by functions such as anova or summary, or by embedding a salient number such as a p-value into the main text.
When presenting your results, you need to give the reader a feeling for the effects that you report. In simple
cases, this can be done either numerically (for example by presenting means with their associated standard errors or confidence intervals or p-values). Where results are too complex for this to be effective, use figures to illustrate important results or patterns. Properly done, figures can represent complicated results very succinctly. But it is key that they are integrated into the main text and complement the message you provide there. This means that they need to be referred to in the main text to illustrate specific points, e.g.,
“as shown in Figure 1, we found that . . .” or “We found that . . . (Figure 1) .” Even though you will refer to
your figures in the text, make sure they have a legend that allows the reader to understand what is shown (but not what it means) by looking at the figure alone.
Key points regarding your report and uploading it:
1. Your report should be a maximum of 1500 words long, including the title and section headings, citations to any scientific articles that you would like to refer to, and the figure and table captions, but not any
bibliography. The word limit applies to the file produced by Knitting the Markdown in RStudio, not
to the length of the Markdown source itself. We will assess word limit by Knitting the file into a Word Document, although we will use the PDF version for marking because it formats Markdown better than Word.
2. Please submit your report as archive (a zip-compressed folder) containing your Markdown source file and the datafile. You can optionally include your compiled PDF if you concerned that knitting the output is unstable.
3. Do not include your name in the Markdown submission: We mark your submissions anonymously which is the fairest thing to do.
In terms of assessment, the emphasis here is not so much on the computation but mostly on generating
and communicating biological insights from data. This is reflected in how we weigh different aspects of your work. We will award 20% of your mark for the correct choice and execution of the statistical analyses, 30%
for the readable and comprehensible presentation of your methods and results using R markdown, 25% for
the effective use of figures and tables (including choosing the appropriate type and number of illustrations),
and 25% for the correct and meaningful interpretation of the results (including any biological conclusions
you draw).
References
Emerson, S.B., 1979. The ilio-sacral articulation in frogs: form and function. Biological Journal of the
Linnean Society, 11(2), pp. 153- 168.
Enriquez-Urzelai, U., Montori, A., Llorente, G.A. and Kaliontzopoulou, A., 2015. Locomotor mode and the evolution of the hindlimb in western Mediterranean anurans. Evolutionary Biology, 42(2), pp.199-209.
James, R.S. and Wilson, R.S., 2008. Explosive jumping: extreme morphological and physiological special- izations of Australian rocket frogs (Litoria nasuta). Physiological and Biochemical Zoology, 81(2), pp.176- 185.
Jorgensen, M.E. and Reilly, S.M., 2013. Phylogenetic patterns of skeletal morphometrics and pelvic traits
in relation to locomotor mode in frogs. Journal of Evolutionary Biology, 26(5), pp.929-943.
Lires, A. I., Soto, I.M. and Gómez, R.O., 2016. Walk before you jump: new insights on early frog locomotion
from the oldest known salientian. Paleobiology, 42(4), pp.612-623.
Soliz, M., Tulli, M.J. and Abdala, V., 2017. Osteological postcranial traits in hylid anurans indicate a mor-
phological continuum between swimming and jumping locomotor modes. Journal of Morphology, 278(3),
pp.403-417.
2023-01-11