IGF-animal-welfare

This repository contains data and source code to reproduce the data and statistical analysis and figures of the manuscript submitted to the journal "Biomarker Research":

Potential of factors of the IGF system as biomarkers for animal welfare assessment by Anne-Marie Galow1, Daniela Ohde1, Anja Eggert1, Christina Walz1, Marianne Zenk1, Chaithra Umesh2, Saptarshi Bej3, Olaf Wolkenhauer2 and Andreas Hoeflich1

1Research Institute for Farm Animal Biology (FBN), Dummerstorf, Germany 2University of Rostock, Dept. of Systems Biology and Bioinformatics, 18051 Rostock, Germany 3IISER Thiruvananthapuram, School of Data Science, 695551 Vithura, Kerala, India

License

This work is licensed under a Creative Commons Attribution 4.0 International License.

In order to reference this software, please consider the information in the CITATION.cff file.

Usage

We run the data and statistical analysis on Windows 11. With one exception, all data processing, modeling and visualization is done in R. Missing value imputation was done in Python.

R Core Team (2024). R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/.

THE REPOSITORY CONTAINS: 1. Main folder with 5 quarto files and rendered reports in html + igf-biomarker-summary-stats.qmd: data processing, data summary statistics, processed data saved in "./data/data-processed.RData" + igf-biomarker-models.qmd: linear mixed models, generating figures, using data-processed.RData + igf-biomarker-pca.qmd: PCA and cluster analysis, using data-processed.RData and imputed data from missing_value_imputation.ipynb + missing_value_imputation.ipynb: Jupyter Notebook with kNN imputation in Python + igf-confounder-age.qmd: linear mixed model including age, supplements + igf-confounder-boar.qmd: linear mixed model including boars, supplements 3. Folder data with one xlsx-files: + data-matrix.xlsx with sheets: + data_ori: measured data + data_ori_knn: subset of variables used for kNN imputation + data_imp: imputed data set + parameter description: some meta information 4. Folder plots with 8 plots: + figure3.png: Effects of the husbandry system on cortisol and IGF bioactivity + figure4.png: Husbandry dependent IGF concentrations + figure5.png: Litter dependent husbandry effects + figure6.png: Husbandry effects on regulators of IGF bioactivity + figure7.png: Principal Component Analysis (PCA) of 15 candidate parameters related to husbandry systems + figure8.png: Dendrogram for hierarchical clustering of imputed dataset + ori-imp.png: Comparison of data distributions for selected candidate parameters of the original and imputed data set, supplement + cluster1-cluster2.png: Comparison of selected candidate parameters for the two main clusters obtained from hierarchical clustering, supplement