Ethanol Impact on Zebrafish Development: Laboratory Protocols and Statistical Analysis

Learning Outcomes and Developmental Context

• The primary objective of the laboratory is to examine ethanol-treated zebrafish embryos, compare them to control embryos, and identify phenotypic differences.

• Key learning objectives include:

  • Accuratly and concisely analyzing and describing class-wide data sets.

  • Critiquing experimental design and identifying the limitations of the laboratory procedures.

  • Relating the observed effects of ethanol on developing embryos to theoretical concepts discussed in lectures.

  • Explaining how perturbations in environmental or genetic regulation of development lead to developmental disorders.

• Ethanol is classified as a teratogen. Investigating its effects on model organisms like chick, zebrafish, and mouse embryos helps clarify the pathology of Fetal Alcohol Spectrum Disorder (Fasd) in humans.

• Zebrafish treated with low doses of ethanol at early developmental stages exhibit phenotypes significantly similar to children born with Fasd.

Experimental Protocol and Subject Observation

• Zebrafish embryos undergo ethanol treatment during a specific developmental window, typically from $4\,hpf$ (hours post fertilization) to $24\,hpf$.

• The embryos are set up on Monday morning and observed at $48\,hpf$ on Wednesday afternoon.

• Observers work in teams of three to ensure data accuracy. Each team member should examine the embryos to reach a consensus on abnormalities before submitting findings to the class data spreadsheet.

• Data collection focuses on two specific, easily observable abnormalities to maintain consistency, as other traits can be difficult for inexperienced observers to identify.

Identification of Morphological Abnormalities

• Cardiac Oedema

  • This refers to swelling or fluid accumulation around the heart.

  • In a normal embryo, the heart sits within a small sac; in affected embryos, this sac becomes visibly enlarged or swollen.

  • Identification requires comparing treated embryos against control embryos to recognize the deviation from normal heart sac size.

• Lordosis

  • This is a skeletal abnormality characterized by a curvature of the spine.

  • Normal zebrafish embryos have a straight, elongated spine.

  • In embryos with lordosis, the spine is bent or curved upward at the tail.

  • An individual embryo may present with both cardiac oedema and lordosis simultaneously.

Class Data Collection and Categorization

• The laboratory seeks to generate a robust data set with at least $50$ embryos per experimental group, though final numbers depend on animal availability.

• Experimental groups include:

  • Control group (no ethanol exposure).

  • $1\%$ ethanol treatment group.

  • $1.5\%$ ethanol treatment group.

• Data is recorded in a categorical table with the following classifications:

  • Number of embryos with cardiac oedema only.

  • Number of embryos with lordosis only.

  • Number of embryos with both cardiac oedema and lordosis.

  • Number of embryos with no abnormalities.

• In addition to morphological data, groups calculate heart rate averages for several embryos to contribute to a pooled class data set for physiological analysis.

Statistical Analysis of Continuous Data (ANOVA)

• Heart rate is considered continuous data because it exists on a spectrum or continuum.

• The goal of statistical analysis is to determine if differences between groups are significant (real) or occurred by chance. This depends on sample size, the magnitude of differences between groups, and the variation within groups.

• One-Way ANOVA (Analysis of Variance)

  • This test is used when comparing more than two groups (e.g., Control, $1\%$, and $1.5\%$ ethanol).

  • Using multiple t-tests instead of an ANOVA increases the risk of a type II error.

  • ANOVA addresses whether at least two groups differ significantly in their average values.

  • It calculates the means of each group relative to the overall average mean and analyzes the variation both within each group and between the groups.

• The F-Statistic and P-Value

  • The F-statistic ($F$) is the ratio of between-group variability to within-group variability.

  • The tool compares the calculated F-statistic to a critical F-value.

  • If the result is significant, a post-hoc test is conducted to identify exactly which groups differ from one another.

  • A significance level of $0.05$ (or $5\%$)) is standard. If the p-value is greater than $0.05$, the null hypothesis is not rejected, meaning variation between groups is not significantly greater than variation within groups.

• Example Analysis

  • In a sample data set where $F = 2.5$ and $p = 0.099$, the result is not statistically significant because the p-value is above the $0.05$ threshold.

Statistical Analysis of Categorical Data (Chi-Square)

• The number of abnormalities is categorized as categorical data because the entries are sorted into discrete groups.

• Chi-Square Analysis

  • This test determines if there is a significant association between categorical variables (e.g., the treatment level and the frequency of abnormalities).

  • The hypothesis tested is that higher ethanol concentrations are associated with a higher frequency of abnormalities.

  • The test compares observed frequencies (the actual data collected) to expected frequencies (calculated by the program based on the data set).

• Procedural Limitations and Alternatives

  • Ideally, a Chi-Square analysis requires each cell in the data table to have a value of $5$ or more. If values are lower (e.g., zero or one in the control group), the Fisher's exact test or a variation of the Chi-Square might be more appropriate.

  • Despite these technical limitations, the standard Chi-Square is used in this class to demonstrate the analytical process.

• Output Metrics

  • The analysis provides a Chi-Square value, degrees of freedom (based on the number of groups and categories), and a p-value.

  • Strength of association can also be measured using Cramer's V ($\phi_c$ or $V$), though this is often more applicable to tests of independence.

  • Reporting significant Chi-Square results requires providing the Chi-Square value and the p-value.