Package Name: SplineOmics
SplineOmics-package.RdThe R package SplineOmics finds the significant features (hits) of time-series -omics data by using splines and limma for hypothesis testing. It then clusters the hits based on the spline shape while showing all results in summary HTML reports.
For detailed documentation, vignettes, and examples, please visit the [SplineOmics GitHub page](https://github.com/csbg/SplineOmics.git).
Key Functions and Classes
- extract_data: Extracts data matrix from Excel file. - create_splineomics: Creates the SplineOmics object, which contains arguments used by several package functions. - explore_data: Performs exploratory data analysis with the data, and outputs an HTML report containg various plots, such as density plots and correlation heatmaps. - screen_limma_hyperparams: Allows the specify lists of different hyperparameters to test, such as a degree of freedom of 2, 3, 4, and adj.p-val thresholds, such as 0.1 and 0.05, and tests all specified different values for all limma spline hyperparameters in a semi-combinatorial way. - update_splineomics: Allows to change values of the SplineOmics object, for example after observing that outliers should be removed from the data (update the data parameter). - run_limma_splines: Central function of the script, is called by the screen_limma_hyperparams function and can be called to get the limma spline analysis results (p-values for all features (e.g. proteins)) with the hyperparameters, that were selected finally. - create_limma_report: Creates an HTML report showing the run_limma_splines results - cluster_hits: Clusters the splines of the hits (significant features) based on their shape and shows all results as plots in an HTML report. - download_enrichr_databases: Allows to download the Enrichr databases for runnin clusterProfiler in the run_gsea function with them. - run_gsea: Runs clusterProfiler with the clustered hits by using the Enrichr databases.
Dependencies
- **ComplexHeatmap**: For creating complex heatmaps with advanced features. - **base64enc**: For encoding/decoding base64. - **dendextend**: For extending `dendrogram` objects in R, allowing for easier manipulation of dendrograms. - **dplyr**: For data manipulation. - **ggplot2**: For creating elegant data visualizations using the grammar of graphics. - **ggrepel**: For better label placement in ggplot2. - **here**: For constructing paths to your project’s files. - **limma**: For linear models for microarray data. - **openxlsx**: For reading, writing, and editing xlsx files. - **patchwork**: For combining multiple ggplot objects into a single plot. - **pheatmap**: For creating pretty heatmaps. - **progress**: For adding progress bars to your loops and apply functions. - **purrr**: For functional programming tools. - **rlang**: For tools to work with core language features of R and R’s base types. - **scales**: For scale functions for visualization. - **tibble**: For creating tidy data frames that are easy to work with. - **tidyr**: For tidying your data. - **zip**: For combining files into a zip file.
Optional dependencies
These dependencies are only necessary for some functions:
- **edgeR**: For preprocessing RNA-seq data in the run_limma_splines() fun. - **clusterProfiler**: For the run_gsea() function (gene set enrichment). - **rstudioapi**: For the open_tutorial() and open_template() functions.
Authors
- [Thomas-Rauter](https://github.com/Thomas-Rauter) - Wrote the package and developed the approach with VSchaepertoens under guidance from nfortelny and skafdasschaf. - [nfortelny](https://github.com/nfortelny) - Principal Investigator, provided guidance and support. - [skafdasschaf](https://github.com/skafdasschaf) - Helped review code and provided improvement suggestions. - [VSchaepertoens](https://github.com/VSchaepertoens) - Developed an internal plotting function and contributed to exploratory data analysis and the overall approach.
Author
Maintainer: Thomas Rauter thomas.rauter@plus.ac.at