Datasets for feature-specific class prediction, from readings in several cell-types.

The cell-types featured here are A549, H1ESC, HeLa, IMR90, K562, and MCF7. These happen to be the 6 cell-types out of 8 existing in our integrative database, GREG (https://doi.org/10.1093/database/baz162). The normalized read-coverages from features (variables) like DNA-Protein bindings, histone modifications; ChIP-Seq data, are mapped through genomic intervals of 2 Kbs, genome-wide. These genomic intervals (technically bins), are classified as Hubs or Non-Hubs. Hubs are those bins that are engaged in genomic crosstalk at a distance of 1Mb or Inf., and are Non-Hubs otherwise.

Note that features are inconsistent across different cell-types, due to the availability of data. The BAM files have been sourced from ENCODE and NCBI repositories. Also, the users might want to use unique() in R or something similar to refine the datasets for non-redundant instances (if existing).

The analysis following this data can be tracked at https://github.com/mora-lab/GREG-Hubs.