A subscription to JoVE is required to view this content. Sign in or start your free trial.

Method Article

Morphometric Analyses of Shape: The Analysis Software Toolbox for Craniofacial Shape Quantification in Zebrafish

929 views

⸱

DOI:

10.3791/70345

⸱

February 27th, 2026

 ,  ,  , 
1Department of Biochemistry and Molecular Genetics, Alcohol Research Center, University of Louisville, 2Department of Pediatrics, Section of Developmental Biology, University of Colorado School of Medicine, Anschutz Medical Campus

In This Article

Summary

This work describes a protocol for quantifying craniofacial cartilage shape using free software (tpsDigs2, MorphoJ, and PAST) to measure changes in facial structure in zebrafish larvae.

Abstract

Fetal alcohol spectrum disorders (FASD) are characterized by a varying set of physical, cognitive, and behavioral disabilities caused by prenatal ethanol exposure, including those affecting the facial skeleton. Ethanol-sensitive genetic loci contribute to this high degree of variation in FASD, which complicates analyses of facial shape. We have previously shown that we can analyze these changes in facial shape from gene-ethanol interactions in zebrafish. Zebrafish are an ideal model to analyze this variation for several reasons; (i) 70% of genes are orthologs between humans and zebrafish; (ii) external fertilization allows researchers to control timing and dosage of ethanol treatments; (iii) the structure of their facial skeletal is conserved with vertebrates; (iv) translucent larvae enable direct viewing of changes to the craniofacial skeletal structure during development. However, analyzing the shape of the craniofacial skeleton can be difficult to fully assess through simple linear measures, as these do not capture overall changes in shape. In addition, changes in head size can complicate data interpretation. To address this, we undertook a morphometric approach, analyzing overall facial shape through principal component analyses via freeware software, tpsDigs2, MorphoJ, and PAST. The combination of this software allows for pairwise comparison of overall facial morphology. Here, we outline our approach and analysis of facial shape in ethanol-treated zebrafish mutants using these programs to conduct a series of complementary multivariate statistical analyses.

Introduction

Fetal Alcohol Spectrum Disorders (FASD) are characterized by a broad range of developmental defects, including behavioral, neurological, and physical1,2. Included in these physical defects are craniofacial defects, such as jaw hypoplasia3,4,5,6. While timing and dosage of ethanol exposure contribute to the complex etiology of FASD, genetic contribution plays a significant role in FASD etiology6,7,8,9,10,11,12. This combination of factors makes studying facial defects in human cohorts challenging. To study the impact of prenatal ethanol exposure on development, we use the zebrafish model. Zebrafish serve as a strong model for FASD as they share 70% gene orthologs with humans, 82% of known disease-causing genes13,14,15,16. In addition to genetic conservation, zebrafish: (i) have high fecundity, allowing for several zebrafish larvae to be produced at a time, (ii) undergo external fertilization, which allows direct study of ethanol-sensitive developmental processes, (iii) have highly conserved and stereotyped vertebrate craniofacial development, and (iv) are translucent embryos/larvae allowing for visualization of skeletal structures17. We have previously shown that mutations in many different zebrafish genes sensitize embryos to ethanol-induced facial defects, and these defects can be subtle and difficult to identify by eye10,18. In addition, we observed that ethanol-treated wild-type larvae also have slight changes in the craniofacial region compared to untreated wild-type larvae, though again these changes are difficult to identify by eye10. Although these changes go undetected upon visual observation, we were able to show these ethanol-induced changes in facial shape using morphometric analyses available on 2D images of the viscerocranium10.

The face is a complex 3D structure that is difficult to measure using conventional linear measurements. These single linear measures do not account for the relative position of each structure measured to, and their impact on, the other measured structures of the face. Morphometric analyses address these shortcomings by using shape configurations via landmarks that account for the relative position of all structures measured, even controlling for overall differences in size18,19,20. This approach results in data that is more accurate, has greater resolution, and is much easier to visualize, which can yield results that may not be observed using conventional linear measurements18,19,20. In addition, morphometric approaches can make direct comparisons between and within groups using multivariate statistical methods. Here we describe the use of the freeware morphometric software, tpsUtil, tpsDigs221, MorphoJ20, and Paleontological Statistics (PAST)22, in combination with 2D images of the viscerocranium in untreated and ethanol-treated wild-type and larvae heterozygous for the Bone Morphogenetic Protein (Bmp) gene bmp7a. While these programs have been utilized mainly in paleontology and ecology-focused fields, when used in combination, they provide a thorough analysis of facial shape. Ultimately, in this protocol, we show how to use these freely available software applications to quantify craniofacial cartilage shape changes in the zebrafish model.

Access restricted. Please log in or start a trial to view this content.

Protocol

All zebrafish larvae used in this procedure were raised and bred following established IACUC protocols approved by the University of Louisville23.

NOTE: The zebrafish strain bmp7aty68a24 and wild-type siblings (all in the AB background) were used in this study, with adult fish maintained at 28.5 °C with a 14/10-h light / dark cycle. Use the appropriate lines per the desired experimental outcomes. Use water that was sterilized by reverse osmosis. Sex as a biological variable does not apply to the studied development stages, as sex is first detectable in zebrafish around 20-25 days post-fertilization (dpf)25, after all of the analyses.

1. Ethanol treatment

  1. Collect eggs from heterozygous crosses and stage embryos morphologically as described23.
  2. Sort embryos into sample groups of 100 and rear at 28.5 °C to desired developmental time points. Incubate all groups in embryo media (EM)10,18.
  3. At 6 hours post fertilization (hpf), change EM to either fresh EM or EM containing 1% ethanol (v/v). At 24 hpf, wash out EM containing ethanol with three fresh changes of EM10,18.

2. Facial staining and imaging

  1. Fix zebrafish larvae at 5 dpf and stain with Alcian Blue, labeling cartilage (Figure 1) as previously described26.
  2. Take whole mount, ventral view, brightfield images of the viscerocranium, mounting larvae as previously described27.
  3. Briefly, mount fish, ventral surface up, in the 50% glycerol solution from the end of Alcian Blue staining.
    NOTE: Settings on compound microscopes can vary; use settings appropriate to image structures desired. An Olympus BX53 compound microscope was used in this study to generate all images analyzed. Images were taken at 10x magnification.

3. Genotyping

  1. Remove the tails from 5 dpf larvae following Alcian Blue cartilage staining as previously described18.
  2. Separate the heads and tails of the larvae. Keep the heads in 50% glycerol storage for later imaging.
  3. Lyse tails with 5 µL of Proteinase K (10 mg/mL) to extract gDNA.
  4. Perform PCR on the gDNA for each sample using primers targeting the gene of interest to amplify regions containing the SNP.
  5. Digest the PCR product using restriction endonuclease-based DNA digestion to differentiate mutant from wild-type alleles based on the SNP to identify the SNP. Divide the samples by genotype into pooled groups and image for analysis.
    NOTE: For this work, the SNP was amplified in bmp7a and genotyped using BslI (NEB).

4. Software used for shape analyses

  1. Use tpsUtil and tpsDig221, and MorphoJ20 (https://sbmorphometrics.org) for morphometric analysis of alcian-stained larvae, and MorphoJ and Paleontological Statistics Software Package for Education and Data Analysis (PAST)22 analysis software (https://www.nhm.uio.no/english/research/resources/past/) for all the statistical analyses. 
    NOTE: The tpsUtil software can be found under the "Utility programs" link, the tpsDig2 software can be found under the "data acquisition" link, and MorphoJ can be found under the "Comprehensive programs" link on https://sbmorphometrics.org. Troubleshooting for each software can be readily found both freely online, through multiple platforms, and within the software. The Principal Component Analysis (PCA), Canonical Variate Analysis (CVA), Procrustes ANOVA, and wireframe graphs of facial variation were generated using MorphoJ, and a multivariate test of normality and a multivariate analysis of variance (MANOVA) were performed using PAST. 

5. Imaging the facial skeleton consistently for all samples to be analyzed

  1. Ensure images of the facial skeleton are taken at identical microscope settings and saved as .tif files.
    NOTE: Microscope settings vary depending on the setup and functionality of the microscope and the samples/tissues to be imaged. The key is a consistent microscope setup and sample preparation.
  2. Save files in a single folder labeled to fully identify the experiment.

6. Preparing images with the tps software

  1. Open Notepad and select Save As. From here, click all files at the bottom. Save the file as "experiment-name.tps" and save it in the folder containing all images to be analyzed.

7. Inputting images into the tps software

  1. Open tpsUtil software. Next, pull down Operation and select the Build tps file from the images tab.
  2. From this tab, click input and open the file containing the images. Click the first image in the file and select Open.
  3. Select the output tab, followed by the tps notepad file, and save. When prompted to replace the file, select yes.
  4. Next, click the set up tab and create. Go to the file to see if the images have been added.

8. Adding landmarks to images

  1. Open tpsDigs2 and open the following tabs in order: Input source, file, then the saved tps Notepad file.
  2. Next, click options to visualize the Image tools option. Here, make sure to set the reference length to 100 µm. Press set scale.
    NOTE: Do not scale relative to the scale on the image.
  3. When setting the scale, press ok to set parameters, then exit image tools.
  4. At this point, add landmarks to each image to be analyzed for proper morphology analysis by clicking the aim symbol.
  5. Ensure landmarks are placed in the same order for each image. Landmarks were placed on the following joints between the cartilage elements (See Figure 1E, E'):
    1. Midline joint between the Meckel's cartilages (Landmark #1, Figure 1E, E').
    2. The bilateral joints between Meckel's and the palatoquadrate cartilages (Landmarks #2, Figure 1E, E').
    3. Midline joint between the ceratohyal cartilages (Landmark #3, Figure 1E, E').
    4. The bilateral joints between the palatoquadrate and ceratohyal cartilages (Landmarks #4, Figure 1E, E').
    5. The distal end of the hyomandibular cartilages (Landmarks #5, Figure 1E, E') .
  6. After placing landmarks on an image, click files for the dropdown menu. Follow this pathway to save data and then overwrite.
  7. Exit tpsDigs2.

9. Using MorphoJ to analyze shape by Procrustes ANOVA

  1. Open the MorphoJ software.
  2. Under preliminaries, create and name the dataset. Select create new dataset and name the dataset. Then, click tps and select the notepad with the new data points added from step 8.4. Create the dataset.
  3. The images are ready to be statistically viewed using Procrustes fit. Select the tab reading project tree and click on the dataset. Select the following buttons in this pathway: preliminaries, new Procrustes fit, Align by principal axes, and Perform Procrustes fit.
  4. Select generate covariance matrix under preliminaries; the software will say Procrustes coordinates. Execute the function when prompted.
  5. Select create or edit wireframe under preliminaries and link the points on the images. Then select link points and accept or create the image. Edit classifiers.
  6. Open a spreadsheet while keeping MorphoJ open. Add all necessary information, for example: Genotype, Treatment, Genotype and Treatment, Experiment, etc. Save the spreadsheet as a CSV file.
  7. In MorphoJ, click file and import classifier variables to select the CSV file. After opening the file, go back to the project tree and click the dataset.
  8. Once the dataset is opened, click preliminaries and edit classifiers. This will ensure all images are added.
  9. Next, click project tree, Covmatrix, then Variation at the top.
  10. Select Principal Component Analysis to see PC scores. Click PC scores to see the generated graph.
  11. To add colors to the graph, right-click and press confidence ellipsis to add the desired classifier. Press color data points to add the colors, and ok to accept these changes.
    NOTE: Once the PC score tab is closed and reopened following the instructions in step 9.10, the colors should be updated.
  12. To change the wireframe colors, go to Preliminaries. Click set options for the shape graph, located at the bottom of the screen. Select wireframe graphs to change the colors of the target shape, starting shape, and numbers.
  13. Select Variation, then Procrustes ANOVA.
  14. Export the Procrustes ANOVA results to the spreadsheet file from step 9.7.

10. Canonical variate analysis of the dataset

  1. In MorphoJ, select the original dataset or project. To begin Canonical Variate Analysis (CVA), select Comparison followed by Canonical Variate Analysis. Select the classifier variable(s) - select Genotype & Treatment and execute the function.
  2. To export CVA, click the results tab and right-click on the now-open results page. Select Export to File and save the information.
  3. Click Project Tree and select CVA, then Scores. At the top left of the page, select File. Once the File tab has opened, choose Export Dataset and select the data type and Genotype & Treatment. Save CVA scores as a .txt file.
  4. To prepare the file for PAST software, open the saved CVA scores in Notepad or an equivalent software. Change ID in the top left corner to Label so as not to confuse PAST software. Save the edited CVA Scores or equivalent.

11. MANOVA in PAST software

  1. To import the CVA scores in PAST, select File, then Open to access the saved CVA scores from Notepad or equivalent. When the Import text file window opens, select Names, Data for row and column, and tab for a separator. Select Import.
  2. Under the Show tab, select Column Attributes. Next to the Type button, open the dropdown menu and select Group for the first column that should have the classifier variable(s).
  3. Highlight the PC data or CV data. Select Multivariate, Tests, and Multivariate Normality to run the normality test. Run this test on PC and CV separately.
  4. Click the gray empty cell in the top left corner, above type, to select the entire dataset. Select Multivariate, Tests, and MANOVA to process the MANOVA. Export the MANOVA results to the spread sheet file from step 9.7.
    NOTE: Both Summary and Pairwise results will be given by the software. Export or save both.

Access restricted. Please log in or start a trial to view this content.

Results

To identify ethanol-induced facial shape changes in zebrafish, we combined freely available software applications to generate and quantify morphological data of the facial skeleton. Embryos from bmp7a heterozygous adult carrier fish were generated and treated with ethanol from 6-24 hpf. Larvae from these crosses were fixed at 5 dpf, and facial cartilages were stained with Alcian Blue. Images of the viscerocranium were taken for each larva in each genotype and treatment group (Figure 1A

Access restricted. Please log in or start a trial to view this content.

Discussion

FASD is characterized by a wide range of developmental defects, including craniofacial defects such as jaw hypoplasia. Zebrafish are a strong model for FASD due to their genetic conservation with humans, high fecundity, translucent larvae, and external fertilization. Zebrafish have been used for decades to study both the formation of the craniofacial skeleton and the impact of ethanol on development 6,12,13,

Access restricted. Please log in or start a trial to view this content.

Disclosures

The authors have nothing to disclose.

Acknowledgements

The research presented in this article was supported by a grant from the National Institutes of Health/National Institute on Alcohol Abuse (NIH/NIAAA) R01AA031043 to C.B.L.

Access restricted. Please log in or start a trial to view this content.

Materials

List of materials used in this article
NameCompanyCatalog NumberComments
tpsUtil, tpsDig2, MorphoJhttps://sbmorphometrics.orgFreeware Software - may require administrative permissions to download
PASThttps://www.nhm.uio.no/english/research/resources/past/Freeware Software - may require administrative permissions to download
Compound Brightfield MicroscopeOlympusBX53

References

  1. May, P. A., Baete, A., Russo, J., et al. Prevalence and characteristics of fetal alcohol spectrum disorders. Pediatrics. 134 (5), 855-866 (2014).
  2. May, P. A., Gossage, J. P., Kalberg, W. O., et al. Prevalence and epidemiologic characteristics of FASD from various research methods with an emphasis on recent in-school studies. Dev Disabil Res Rev. 15 (3), 176-192 (2009).
  3. Klingenberg, C. P., Wetherill, L., Rogers, J., et al. Prenatal alcohol exposure alters the patterns of facial asymmetry. Alcohol. 44 (7-8), 649-657 (2010).
  4. Smeriglio, P., Zalc, A. Cranial neural crest cells contribution to craniofacial bone development and regeneration. Curr Osteoporos Rep. 21 (5), 624-631 (2023).
  5. Smith, S. M., Garic, A., Flentke, G. R., Berres, M. E. Neural crest development in fetal alcohol syndrome. Birth Defects Res C Embryo Today. 102 (3), 210-220 (2014).
  6. Vo, H. D. L., Lovely, C. B. Ethanol induces craniofacial defects in Bmp mutants independent of nkx2.3 by elevating cranial neural crest cell apoptosis. Biomedicines. 13 (3), (2025).
  7. Brooks, E. R., Islam, M. T., Anderson, K. V., Zallen, J. A. Sonic hedgehog signaling directs patterned cell remodeling during cranial neural tube closure. Elife. 9, e60234(2020).
  8. Eberhart, J. K., Parnell, S. E. The genetics of fetal alcohol spectrum disorders. Alcohol Clin Exp Res. 40 (6), 1154-1165 (2016).
  9. Fish, E. W., Murdaugh, L. B., Sulik, K. K., Williams, K. P., Parnell, S. E. Genetic vulnerabilities to prenatal alcohol exposure: Limb defects in sonic hedgehog and GLI2 heterozygous mice. Birth Defects Res. 109 (11), 860-865 (2017).
  10. Klem, J. R., Schwantes-An, T. H., Abreu, M., et al. Mutations in the bone morphogenetic protein signaling pathway sensitize zebrafish and humans to ethanol-induced jaw malformations. Dis Model Mech. 18 (4), dmm052223(2025).
  11. Swartz, M. E., Lovely, C. B., McCarthy, N., Kuka, T., Eberhart, J. K. Novel ethanol-sensitive mutants identified in an F3 forward genetic screen. Alcohol Clin Exp Res. 44 (1), 56-65 (2020).
  12. Swartz, M. E., Wells, M. B., Griffin, M., et al. A screen of zebrafish mutants identifies ethanol-sensitive genetic loci. Alcohol Clin Exp Res. 38 (3), 694-703 (2014).
  13. Fernandes, Y., Lovely, C. B. Zebrafish models of fetal alcohol spectrum disorders. Genesis. 59 (11), e23460(2021).
  14. Howe, K., Clark, M. D., Torroja, C. F., et al. The zebrafish reference genome sequence and its relationship to the human genome. Nature. 496 (7446), 498-503 (2013).
  15. Lovely, C. B. Animal models of gene-alcohol interactions. Birth Defects Res. 112 (4), 367-379 (2020).
  16. Lovely, C. B., Fernandes, Y., Eberhart, J. K. Fishing for fetal alcohol spectrum disorders: Zebrafish as a model for ethanol teratogenesis. Zebrafish. 13 (5), 391-398 (2016).
  17. Marrs, J. A., Sarmah, S. The genius of the zebrafish model: Insights on development and disease. Biomedicines. 9 (5), 577(2021).
  18. Gray, R., Lloyd, A., Lovely, C. B. Interactions between the bone morphogenetic protein and the planar cell polarity pathways lead to distinctive ethanol-induced facial defects. Alcohol Clin Exp Res (Hoboken). 49 (11), 2502-2519 (2025).
  19. Mutsvangwa, T., Douglas, T. S. Morphometric analysis of facial landmark data to characterize the facial phenotype associated with fetal alcohol syndrome. J Anat. 210 (2), 209-220 (2007).
  20. Klingenberg, C. P. MorphoJ: An integrated software package for geometric morphometrics. Mol Ecol Resour. 11 (2), 353-357 (2011).
  21. Rohlf, F. The Tps series of software. Hystrix. 26, 1-4 (2015).
  22. Hammer, O., Harper, D., Ryan, P. PAST: Paleontological statistics software package for education and data analysis. Palaeontol Electron. 4, 1-9 (2001).
  23. Westerfield, M. The Zebrafish Book. A Guide for the Laboratory Use of Zebrafish (Danio rerio). , 5th edition, University of Oregon Press. Eugene. (2007).
  24. Dick, A., Hild, M., Bauer, H., et al. Essential role of Bmp7 (snailhouse) and its prodomain in dorsoventral patterning of the zebrafish embryo. Development. 127 (2), 343-354 (2000).
  25. Aharon, D., Marlow, F. L. Sexual determination in zebrafish. Cell Mol Life Sci. 79 (1), 8(2021).
  26. Walker, M. B., Kimmel, C. B. A two-color acid-free cartilage and bone stain for zebrafish larvae. Biotech Histochem. 82 (1), 23-28 (2007).
  27. McGurk, P. D., Lovely, C. B., Eberhart, J. K. Analyzing craniofacial morphogenesis in zebrafish using 4D confocal microscopy. J Vis Exp. (83), e51190(2014).
  28. Lovely, C. B. Quantification of ethanol levels in zebrafish embryos using head space gas chromatography. J Vis Exp. (156), e60766(2020).
  29. Klem, J. R., Gray, R., Lovely, C. B. The zebrafish Tol2 system: A modular and flexible gateway-based transgenesis approach. J Vis Exp. (189), e64679(2022).
  30. Suttie, M., Foroud, T., Wetherill, L., et al. Facial dysmorphism across the fetal alcohol spectrum. Pediatrics. 131 (3), e779-e788 (2013).
  31. Lovely, C. B., Swartz, M. E., McCarthy, N., Norrie, J. L., Eberhart, J. K. Bmp signaling mediates endoderm pouch morphogenesis by regulating Fgf signaling in zebrafish. Development. 143 (11), 2000-2011 (2016).
  32. Sheehan-Rooney, K., Swartz, M. E., Lovely, C. B., Dixon, M. J., Eberhart, J. K. Bmp and Shh signaling mediate the expression of satb2 in the pharyngeal arches. PLoS One. 8 (3), e59533(2013).
  33. Dong, X. R., Wan, S. M., Zhou, J. J., et al. Functional differentiation of BMP7 genes in zebrafish: bmp7a for dorsal-ventral pattern and bmp7b for melanin synthesis and eye development. Front Cell Dev Biol. 10, 838721(2022).
  34. McCarthy, N., Wetherill, L., Lovely, C. B., Swartz, M. E., Foroud, T. M., Eberhart, J. K. Pdgfra protects against ethanol-induced craniofacial defects in a zebrafish model of FASD. Development. 140 (15), 3254-3265 (2013).

Access restricted. Please log in or start a trial to view this content.

Reprints and Permissions

Tags

Morphometric AnalysisZebrafish ModelFetal Alcohol SpectrumLandmark-Based MorphometricsPrincipal Component AnalysisProcrustes ANOVACanonical Variate AnalysisMultivariate Analysis
Morphometric Analyses of Shape: The Analysis Software Toolbox for Craniofacial Shape Quantification in Zebrafish
Contact Us
Recommend to Library

Privacy

Terms of Use

Policies

Copyright © 2026 MyJoVE Corporation. All rights reserved.