Here are some features of the Omics Playground. You will be able to use them yourself because our platform makes it really easy for anyone to visualize, analyze and explore your big omics data without any coding. Discover hidden patterns and trends in your data effectively. Our goal is to make omics data analysis enjoyable and accessible.
Expression and enrichment analysis
For differential expression, Omics Playground resorts to up to eight different statistical testing methods. For gene set enrichment analysis, we use more than 50’000 gene sets and up to seven different statistical methods. We combine multiple statistical methods using meta-analysis, so you can really trust your results. We also perform drug enrichment analysis accessing a database of more than 5’000 drugs, and find drugs (alone and in combination) with similar or opposing signatures compared to yours. You can enter custom gene sets to perform enrichment analysis of your own signature.
Variable importance and biomarker selection
To understand which genes, mutations or gene sets are more influential on the outcome of your study, the Playground platform calculates variable importance using state-of-the-art machine learning algorithms. The outcome can be multiple categories (classes) or patient survival. Using multiple algorithms we select the best biomarkers. We then train classifiers to build a prediction model for your data and phenotype of interest.
Special immuno-oncology analytics
We have implemented a special analysis modules for immuno-oncology. You can infer copy number variation from single-cell RNA-seq expression, analyze differential immuno-genes usage and perform computational immune celltype profiling using state-of-the-art deconvolution algorithms.
Advanced batch effects analysis
Omics data are notoriously sensitive to so-called “batch effects” that can make or break your analysis. Within the Playground you can easily identify batch effects with our unique PCA cluster analysis. Quantify batch effects and visually evaluate a variety of batch correction techniques in real time.
Systems biology and multi-omics integration
Biology cannot be explained using a single omics data type only. To truly understand biology, we need to measure multiple omics data types and integrate heterogeneous information at gene, functional and phenotype level. Our multi-omics module analyzes gene expression, methylation, copy number and mutation data, and integrates these data using our proprietary multilayer graph algorithm. The result is a better systems biology analysis of your data that provides improved biomarker discovery, functional analysis and outcome prediction.