Describing the Mechanisms Underlying Individual Variation in Coping Styles.
Wednesday, July 22, 2015
Nobody better than our customers themselves to talk about their experience with AutoDiscovery. At the last EAS Aquaculture Congress, Dr Sonia Rey-Planellas and Dr Simon Mackenzie presented their conclusions on the workpackage TRANSCOPE within the framework of the COPEWELL project.
Would you like to know more?
The COPEwell project
The COPEwell project aims is to develop a new integrative framework for the study of fish welfare based on the concepts of allostasis, appraisal and coping styles.
It wants to provide a deeper understanding of the underpinning mechanisms involved in variation in individual coping styles and ability. The project also focuses on the understanding of how fish experience their world, and what effects early life experiences have on later development and coping abilities.
The aim of the TRANSCOPE work package is to integrate and explore the links between natural variation in brain gene expression, behaviour and adaptive physiology, especially aiming to understand how fish species like Atlantic salmon, sea bass and sea bream develop and maintain the ability to cope with stressors, and how behavioural phenotypes are distributed within aquaculture populations.
Our exploratory questions
To properly drive this analysis process, we defined the following exploratory questions:
Which genes are most informative in our subpopulation of Salmon?
Which influence has the tank on fish population total weight and gene copy number?
Exploratory data analysis with AutoDiscovery
Data on gene expression for each individual was correlated with the corresponding behavioural data already obtained in coordinated WPs within COPEWELL. Identification of key brain regions of activity in respect to differential gene expression correlated to coping styles (for all target species, D1.9, M48) is currently underway.
Exploratory analysis of the gene expression data in relation to the behavioural data was performed with AutoDiscovery. It evaluated the Spearman’s Rank correlation coefficients for every pair of numerical variables and one-way ANOVAs for every pair of qualitative-numerical variables within the consolidated data set in order to extract the most relevant relationships between the variables.
Correlations with behavioral data (Hypoxia test sample 1) and between all mRNAs gene expression tank, sex and weight were also performed.
Data was tested for normality with a Kolmogorov-Smirnoff test and Levene’s test for homogeneity of variances.
Tank effects on fish population total weight and gene copy number were also checked either with paired samples T-test or ANOVA test performed by AutoDiscovery along with SPSS. Post-hoc Scheffé or Dunnett test, for non-homogeneous variances, were performed for specific significances.
A GLM ANCOVA was used to test for significant differences in gene expression between all individuals screened for stress coping styles on the population studied. Dependent variable was Log copy number for each target gene with coping style and treatment (when applied) as the fixed factors. Tank and weight were used as covariables (when relevant and depending on the species).
Answering the exploratory questions
The Hypo Booster tool helped us to identify the most informative target genes and also to better understand the influence of the tank on the weight and mRNA copy number variables.
The most informative genes
This exploratory analysis gave us results on correlations between gene expression where PTBP and CRY mRNA copy number where highly correlated. Also correlations were found between PTBP and IFRD1 mRNA copy number.
This means that PTBP in this group of genes doesn’t give us very much information with respect to the variation in gene expression.
The tank effect
Our subpopulation of Salmon data showed a clear influence of tank on fish Total weight. Gene copy number was independent of brain weight but there was also a tank effect.
This figure shows the effect of tank pertinence on mRNA copy number. Tank was used as a covariable for successive analysis on gene expression (p<0.001).