Understanding genome-wide regulation of cell decisions and functions is essential for establishing mechanisms of development, disease, and complex traits. Yet, causal inference remains challenging as >90% of disease-associated genetic variants map to non-coding regions of the human genome, and current methods for variant prioritization are often limited to annotated elements like protein-coding genes. Here, we analyzed 833 epigenomes from the EpiMap repository to identify distinguishing features of genome regulation associated with cell-type specification. We show that the a priori deposition of genome-wide H3K27me3 across diverse cell-types, which we call a repressive tendency score (RTS), provides an approach to enrich for coding and non-coding features of the genome governing cell identity. While only 1.4% of the human genome have significant RTS values, these prioritized regions are enriched in genes, non-coding RNAs, cis-regulatory elements and SNPs associated with fundamental regulation of cell differentiation, organ morphogenesis and disease. We tested the ability to prioritize causal genetic variants contributing to complex traits and diseases by interfacing genome-wide RTS values with data on human genetic variation. Analysing millions of genetic variants from databases including the UK Biobank and ClinVar, we offer an unsupervised approach to rank variant pathogenicity in a disease-agnostic manner to prioritize causal variants, particularly contributors to congenital disorders. Further, we use the RTS metric to systematically predict >8000 cell-type specific cis-regulatory elements, including novel enhancers, silencers, and bifunctional regulatory elements to expand our understanding of regulatory control of cell-type specification. We validate the predictive power of our computational method by analysing CAGE-seq data to identify and experimentally validate a lncRNA, RP11-175E9.1, as a novel regulator of hiPSC differentiation in vitro. Together, this study demonstrates that the conservation of epigenetic regulatory logic provides an effective strategy for studying any omics data-type to infer cell-type specific mechanisms that underpin molecular regulation of cell identity.