Measures of general deleteriousness, like CADD or PolyPhen, have become indispensable tools in the interpretation of genetic variants. However, these estimates say little about where in the organism these deleterious effects will be most apparent. An additional, complementary measure is needed to link deleterious variants (as determined by e.g., CADD) to tissues in which their effect will be most meaningful. Here, we introduce TiSAn (Tissue Specific Annotation), a tool that predicts how related a genomic position is to a given tissue (http://github.com/kevinVervier/TiSAn). TiSAn uses machine learning on genome-scale, tissue-specific data to discriminate variants relevant to a tissue from those having no bearing on the development or function of that tissue. Predictions are made genome-wide, and these scores can then be used to contextualize and filter variants of interest in whole genome sequencing or genome wide association studies (GWAS). We demonstrate the performance and versatility of TiSAn training predictive models for human heart and brain, and detecting tissue-relevant variations in large cohorts for autism spectrum disorder (using TiSAn-brain) and coronary artery disease (using TiSAn-heart). We find that TiSAn is better able to prioritize genetic variants according to their tissue-specific action than the current state of the art method, GenoSkyLine.