refine.bio Compendia packages the data processed by refine.bio pipelines for flexible and broad use by computational biologists and data scientists

The compendium is created by aggregating all samples from a species, removing samples and genes with too many missing values, imputing missing values with SVD impute. The final step is quantile normalization.

Data scientists and computational biologists can use species compendia to quickly construct training and testing sets and extract features from a diverse set of biological contexts and conditions.