RSAT suite

RSAT suite provides computational methods to identify and analyze regulatory signals in non-coding DNA sequences, including motif discovery, genome-scale pattern matching, and statistical assessment of oligonucleotide over-representation.

Key Features:

• Pattern Discovery and Matching: Implements string-based and matrix-based methods for discovering known and de novo regulatory patterns and detecting over-represented oligonucleotides, with multiple pattern-discovery algorithms.
• Genome-Scale Analysis: Performs genome-scale pattern matching using a database of over 100 fully sequenced genomes updated from GenBank, NCBI, and Ensembl.
• Feature Map Drawing: Generates feature maps that annotate positions of identified regulatory patterns.
• Random Sequence Generation: Produces random sequences using Bernoulli and Markov chain background models for control experiments and statistical assessment.
• Web Services Integration: Exposes programmatic web services for integration into automated analysis workflows.
• Phylogenetic Footprint Detection: Provides matrix-based detection of cis-acting elements with adaptive background models and Markov-chain estimation of P-values.

Scientific Applications:

• Transcription factor binding site identification: Detects DNA binding sites for transcription factors across genomes, including Saccharomyces cerevisiae.
• Regulatory element prediction: Predicts unknown regulatory elements and motifs from non-coding sequences.
• Motif validation: Assesses and validates experimentally discovered motifs using statistical significance derived from oligonucleotide frequencies.
• Comparative regulatory analysis: Supports genome-wide and cross-organism analyses of regulatory sequences using available sequenced genomes.

Methodology:

Performs exhaustive oligonucleotide analysis focused on over-represented sequences in upstream regions of co-regulated genes; applies string-based and matrix-based pattern detection; uses adaptive background models and Markov-chain estimation of P-values; generates random sequences with Bernoulli and Markov chain models and defines significance from observed oligonucleotide frequencies across non-coding yeast genome sequences.