regBase | wise aggregation and functional prediction

regBase is a Python library. regBase has no bugs, it has no vulnerabilities, it has a Permissive License and it has low support. However regBase build file is not available. You can download it from GitHub.

regBase is a comprehensively integrated non-coding regulatory prediction scores and composite prediction models from existing tools for base-wise annotation of human genome. As such, the regBase resource provides convenience to prioritize functional regulatory SNVs and assist the fine mapping of causal regulatory SNVs without queries from numerus sources. Inspired by the evident significance of ensemble prediction for pathogenic/deleterious nonsynonymous substitution, regBase also systematically constructs three composite models to score functional, pathogenic and cancer driver non-coding regulatory SNVs. The prediction scores of these models are also intergrated into regBase dataset. - regBase V1.1.1 is released. Three genome-wide functional scores were added into regBase database, including CADD v1.4, CADD v1.6, CScape-Somatic. Now regBase contains 21 tools. Also, regBase database now is stored chromosome by chromosome for easy downloading. regBase Common (see V1.1) and regBase prediction (see V1.0) remain unchanged in this version. - regBase V1.1 is released. Five genome-wide functional scores were added, including DVAR, FitCons2, ncER, Orion, PAFA. Now regBase contains 18 tools and regBase Common contains 28 tools. - regBase V1.0 is released. Its current version is compiled from 23 different tools on functional annotation of non-coding variants, including Basset, CADD, CATO, CDTS, CScape, DANN, DanQ, DeepSEA, deltaSVM, Eigen, FATHMM-MKL, FATHMM-XF, FIRE, fitCons, FunSeq, FunSeq2, GenoCanyon, GWAS3D, GWAVA, LINSIGHT, ReMM, RSVP, SuRFR and PRVCS. Since some tools only support annotations for 1000 Genome Project variants, or take long runtime to compute functional scores, it first built a database, called regBase Common, which contains functional scores from 23 tools for 38,248,779 in the 1000 Genome Project phase 3. Among these integrated dataset, 13 tools provide precomputed scores for almost all possible substitutions of single nucleotide variant (SNV) in the human reference genome. Therefore, it also constructed a complete base-wise aggregation of non-coding variant functional scores for 8,575,894,770 substitutions of SNV, called regBase.