Motif database & search¶

scanRBP ships with a database of RNA-binding protein PWMs, built from two sources:

mCross — Feng H, Bao S, et al., Modeling RNA-Binding Protein Specificity In Vivo by Precisely Registering Protein-RNA Crosslink Sites, Molecular Cell, 2019. Motifs derived from eCLIP crosslink sites.
CISBP-RNA — Ray D, Kazan H, Cook KB, Weirauch MT, Najafabadi HS, et al., A compendium of RNA-binding motifs for decoding gene regulation, Nature, 2013.

Entry naming¶

Each database entry has a scan_id of the form <PROTEIN>.<CELL_LINE_OR_TISSUE>.<cluster>, for example HLTF.K562.01 or DDX3X.HepG2.01 — the same protein can appear multiple times if it was profiled in more than one cell line or has more than one binding cluster.

Searching the database¶

Use the search command to find which entries match a protein name, alias, or description:

scanRBP search TARDBP

[scanRBP] Found proteins in the scanRBP database:
scan_id           protein  tissue  description                                source
TARDBP.K562.00    TARDBP   K562    TAR DNA binding protein                   mCross
TARDBP.HepG2.00   TARDBP   HepG2   TAR DNA binding protein                   mCross

The search matches against the scan_id, protein name and description (case-insensitive substring match), so a search for an alias or partial name still finds the right entries. The -protein scoring option (see Scoring sequences) uses this same matching, so it's worth searching first to see exactly which scan_ids a name will resolve to.

Where the database lives¶

On first use, scanRBP downloads the database (mCross + CISBP-RNA PWMs, ~6 MB) to ~/scanRBP_data. Configuration is stored in ~/.scanRBP (created automatically), with a single setting:

data_folder="~/scanRBP_data"

To move the data folder, run:

scanRBP config /path/to/data_folder

This updates ~/.scanRBP and re-downloads the database there on next use if it isn't already present.

Other resources (not yet integrated)¶

Datasets under consideration for future additions to the database, kept here for reference:

Additional PWM datasets

RCRUNCH — motif compendium, with a data file of candidate PWMs.

CLIP datasets (for use with CLIP-based scoring rather than the PWM database)

ENCODE eCLIP experiments — released eCLIP experiments in K562/HepG2.

Gene annotation

NCBI gene metadata (names, aliases) — the source used to build the aliases/description fields in the scanRBP database.