Morphometric Analysis of Surface Utricles in Halimeda tuna (Bryopsidales, Ulvophyceae) Reveals Variation in Their Size and Symmetry within Individual Segments
Round 1
Reviewer 1 Report
The manuscript provides interesting data on the size, shape and asymmetry of surface urticles. It provides interesting multilevel approach between the surface tesselation, the packing of the utricles and the size and shape of the segments in relation to their position within the segment. The methods are well described and appropriate.
Don't you think that the size gradient (L340) could be related to urticle packing within basally constricted shape ? Across the manuscript, you don't really link the variation in segment shape to the variation of size and shape of utricles. Similarly the variation in symmetry of utricles (L382) seems to agree to the same patterns but no link is apparent.
Also, in the result, the n-gon distribution was provided but it was not used afterwards (does the symmetry vary across those different type of polygons ? Are they distributed evenly or also their density differs between central and marginal position?). This also relates to utricle's packing.
Minor comments:
L63. in my opinion "quantitative" is not necessary
L64-65 the fact to give the name of the species' authors within parenthesis followed by the genus' author doesn't sound a classic way and is pretty confusing to me.
L67. I think it's Fig. 1
L147. explain what are the P_i
Author Response
1) The manuscript provides interesting data on the size, shape and asymmetry of surface urticles. It provides interesting multilevel approach between the surface tesselation, the packing of the utricles and the size and shape of the segments in relation to their position within the segment. The methods are well described and appropriate.
Response: Thank you for this positive assessment of the study. We are really grateful for this encouragement.
2) Don't you think that the size gradient (L340) could be related to urticle packing within basally constricted shape ? Across the manuscript, you don't really link the variation in segment shape to the variation of size and shape of utricles. Similarly the variation in symmetry of utricles (L382) seems to agree to the same patterns but no link is apparent.
Response: Yes, this is an interesting point. Utricle packing is certainly a key mechanism leading to their arrangement in an irregular polygon mesh. From a biological point of view it is also a necessary prerequisite for isolation of the internal segment space and, thus onset of calcification. Following the rationale of your point, smaller size of the basally located utricles might have been related to closer packing in these parts of segments. This would probably mean that these utricles should be longer to account for the smaller area of their apical (surface) polygonal part. Such variation in utricle length has not yet been reported in the literature but it is certainly a possibility and may be teste in the future.
Following your point we have supplemented a new part into the "Discussion" of the revised version that specifically mentions differential packing of utricles within segments that might lead to a smaller size of their polygonal surface faces in the segment bottom parts (l. 368-372).
Relation of segment shape and utricle parameters (size and symmetry) has been evaluated in part 3.5. of the "Results" (Morphology of segments and their relation to utricle size and symmetry) in lines 336-344 and in Table S5. In fact, most of the utricle parameters proved to be unrelated to segment shape patterns. Relation of the single utricle parameter that was significantly related to segment shapes (mean utricle size in basal parts of segments) is illustrated in Figure 4c. Given the fact that other utricle parameters did not yield significant relationships with segment shapes we did not spent too much text for this part of the study in the "Discussion". However, it is clear that it should have been more properly mentioned. In the revised version, two paragraphs of the "Discussion" now explicitly focus on these results (l. 417-425).
3) Also, in the result, the n-gon distribution was provided but it was not used afterwards (does the symmetry vary across those different type of polygons ? Are they distributed evenly or also their density differs between central and marginal position?). This also relates to utricle's packing.
Response: Oh yes, thank you very much for this point. This is clearly something that should have been evaluated and discussed more properly in the original version of the manuscript. In the revised version, we supplemented the analysis of the sample variance of the n-gon distributions in individual positions based on the comparison of the 95% confidence intervals created by the bootstrapping of the original data. In addition, we now also compare the proportions of hexagons in individual datasets. These analyses indeed show that higher symmetry of the centrally located polygons was accompanied by higher propotion of hexagons and lower sample variance of the n-gon distribution in polygons sampled from this parts of segments in comparison with the marginal positions. The respective parts were supplemented in Material and Methods (l. 169-172), Results (l. 297-306) nd Discussion (l. 401-402).
4) L63. in my opinion "quantitative" is not necessary.
Response: Corrected accordingly (l. 63).
5) L64-65 the fact to give the name of the species' authors within parenthesis followed by the genus' author doesn't sound a classic way and is pretty confusing to me.
Response: Actually, this is the standard way how the authors of the taxonomic names are depicted in botany, i.e. according to the botanical nomenclatoric code. The names in the parenthese refer to those authors that originally described and typified the species "tuna", although they classified it into a different genus. Then the name after the parentheses refers to the author of the re-classification of this species into the genus Halimeda.
6) I think it's Fig. 1
Response: Yes, it was supplemented accordingly (l. 67).
7) L147. explain what are the P_i
Response: It was supplemented accordingly (l. 149).
Reviewer 2 Report
This paper deals with morphometrics evaluation of the utricles in segments of green macroalgae.
The study uses measures of area, symmetry and shape (number of polygon vertices).
Evaluation was performed on individual utricles at different positions on the plants segments
and on several segments per plant. Analysis was performed on the measured data to reveal dependencies and relationships between the measured factors and the position within the segment as well as the area and symmetry measures of the segments themselves. Authors discuss the possible relationship between the found morphometric characteristic and the
plant's growth process.
This work is not in my field of expertise and so , I cannot comment on the novelty nor the contribution of this
work to the field. I restrict my comments to the quantification and measures.
The authors present the description and details of the CSM and the Procrustes analysis, however
these can be proven to be mathematically related (as mentioned in paper) and thus I do not see the point
nor interest to the readers in mentioning the two. Same goes for centroid size and polygon area.
In line 148-50-
Authors state that the CSM values are in the range 0 to 1. However this is true ONLY if the points/shape
have been normalized (e.g. so that the CS val as defined in paper is a constant and equal to 1). Normalization is
not mentioned in paper. If normalization is not performed then the CSM value is strongly affected by shape size.
Authors should verify this point in data as well as in text.
* Line74 : "...seed... is located beneath the surface..."
Voronoi seeds are always in the same plane as the voronoi tesselation.
* Line 157 "The mean shape configuration...." Not clear
If I understand correctly, the plant segments are in a linear sequence. In this case another possible
parameter might be the position of the segment within the plant.
In the discussion, the authors discuss the possible relationship between the found measures and the
developmental process of the segment. Is there a similar insight re the parameters of the segments and the
growth process of the segments themselves?
The authors measured symmetry and shape of individual utricles. I think it would be interesting to
evaluate symmetry of the polygonal mesh rather than the individual. (basically translational symetry).
See studies by Yanxi Liu for example
https://www.researchgate.net/publication/6451387_A_Lattice-Based_MRF_Model_for_Dynamic_Near-Regular_Texture_Tracking
or
http://vision.cse.psu.edu/publications/publications.shtml
Author Response
1) This paper deals with morphometrics evaluation of the utricles in segments of green macroalgae. The study uses measures of area, symmetry and shape (number of polygon vertices).
Evaluation was performed on individual utricles at different positions on the plants segmentsand on several segments per plant. Analysis was performed on the measured data to reveal dependencies and relationships between the measured factors and the position within the segment as well as the area and symmetry measures of the segments themselves. Authors discuss the possible relationship between the found morphometric characteristic and the plant's growth process. This work is not in my field of expertise and so , I cannot comment on the novelty nor the contribution of this work to the field. I restrict my comments to the quantification and measures.
Response: Nevertheless, thank you for your cogent and useful points to individual methodological parts of the study.
2) The authors present the description and details of the CSM and the Procrustes analysis, however these can be proven to be mathematically related (as mentioned in paper) and thus I do not see the point nor interest to the readers in mentioning the two. Same goes for centroid size and polygon area.
Response: Yes, this is true. Therefore, we only included the results of Procrustes analysis of polygons and their CS values in the supplementary data and not in the main body of the text. Close mathematical relationship of CSM and PDs and polygon area with CS should have been mentioned - and we have now supplemented it in the revised version (l. 239-240).
3) In line 148-50- Authors state that the CSM values are in the range 0 to 1. However this is true ONLY if the points/shape have been normalized (e.g. so that the CS val as defined in paper is a constant and equal to 1). Normalization is not mentioned in paper. If normalization is not performed then the CSM value is strongly affected by shape size. Authors should verify this point in data as well as in text.
Response: Oh yes, of course. The coordinates of individual vertices were normalized following the original protocol of Zabrodsky et al., i.e. the size of each polygon was scaled so that the longest distance from its center of mass to any of the vertices is one. This necessary information was supplemented into the revised text (l. 142-144) and it is also illustrated in Figure 1d.
4) Line74 : "...seed... is located beneath the surface..."
Voronoi seeds are always in the same plane as the voronoi tesselation.
Response: The sentence was corrected accordingly (l. 73-74).
5) Line 157 "The mean shape configuration...." Not clear
Response: What was meant in this sentence is simply the mean configuration yielded by the generalized Procrustes analysis of the cyclic symmetry group of n-th order based on symmetric transformations of an analyzed n-gon. The sentence was supplemented and rephrased accordingly (l. 158).
6) If I understand correctly, the plant segments are in a linear sequence. In this case another possible parameter might be the position of the segment within the plant.
Response: This is an interesting point. Relation of the serial sequence of segments and their morphology was studied in several previous studies (Neustupa & Nemcova 2018, Plos One; Verbruggen et al. 2005, Crypt. Algol). The key feature of this relationship is higher proportion of relatively small segments with shape characteristics deviating from the species-specific patterns of comparatively larger segments in lower portions of thalli, i.e. close to their base. In the context of the present study this means that the size gradient of utricles within segments is probably more pronounced in segments usually occuring in higher parts of the thalli and vice versa. This information was supplemented into "Discussion" (l. 380-381).
7) In the discussion, the authors discuss the possible relationship between the found measures and the developmental process of the segment. Is there a similar insight re the parameters of the segments and the growth process of the segments themselves?
Response: In fact, so far there aren't any studies focused on an analysis of the segment morphogenesis in relation to their final shapes. Thus, any morphogenetic explanation of the observed plasticity is still missing. The studies published so far were focused on description of the observed variation and on an analysis of the shape variation in relation to the position within the series of segments (see the point above). In this respect, the present study is a pioneering attempt to link the variation in surface utricles to segment shape features. This is one of the things that should make this study an interesting and useful piece of science that will significantly contribute to the progress in current understanding of the patterns and causes of phenotypic plasticity of these important marine algae.
8) The authors measured symmetry and shape of individual utricles. I think it would be interesting to evaluate symmetry of the polygonal mesh rather than the individual. (basically translational symetry). See studies by Yanxi Liu for example https://www.researchgate.net/publication/6451387_A_Lattice-Based_MRF_Model_for_Dynamic_Near-Regular_Texture_Tracking or http://vision.cse.psu.edu/publications/publications.shtml
Response: Thank you for this point and for these tips. Obviously, these studies point to one of the promising and important directions for future research. In this respect, it might be possible to develop the automated registration protocols for the 2D polygon structure of Halimeda segments based on the pattern recognition techniques preented in the studies mentioned in your point. This might allow registration of entire polygon tessellations that might then be used for quantification of their size or symmetry. Thus, progress in this direction might significantly accelerate data acquisition for this kind of research. Following this comment, we supplemented a new part into "Discussion" mentioning a promise for the methodological progress of this kind (l. 454-458). The additional reference [46] was supplemented into the list (l. 574-575).
