Quick Start

Scoring a single sequence

Give scanRBP a sequence directly on the command line, plus an output name:

scanRBP AAAGCGGCGACTTATTATATCCCCATATATTATATCTTCTTCTCTTATATATAAACCAGAGATAGATGTGTGTGGTGG example1 -heatmap example1

This scores the sequence against every RBP motif in the database and writes:

• example1.tab.gz — the log-odds score matrix (proteins × positions), gzipped TSV.
• example1.png / example1.pdf — a clustered heatmap of the scores (only produced when -heatmap is given).

Scoring a FASTA file

Point scanRBP at a FASTA file instead, and it scores every sequence in it, producing one matrix (and, with -heatmap, one heatmap) per sequence, named after that sequence's FASTA id:

scanRBP data.fasta -heatmap data

Reading the heatmap

Rows are RBPs (clustered by similarity of their binding profile across the sequence), columns are sequence positions. Warmer cells mean a stronger predicted binding signal at that position for that RBP; cooler/negative cells mean the position looks less like that RBP's motif than background.

Next steps

• Motif database & search — which RBPs are available, and how to search for one by name.
• Scoring sequences — the full set of options: limiting to specific proteins, heatmap styling, cumulative plots across many sequences.
• CLIP-based scoring — score against real binding evidence instead of a PWM.
• Python API — call scanRBP from your own Python code.