SULDEX

SULDEX measures transcription factor (TF) binding affinities across large DNA sequence repertoires using high-throughput sequencing and hierarchical Bayesian inference.

Key Features:

• High-Throughput Sequencing: Analyzes thousands of potential DNA ligands by sequencing both pre-bound and bound target pools.
• Simultaneous Quantitative Analysis: Quantitatively compares TF binding degrees across tens of thousands of putative binding targets in a single experiment.
• Hierarchical Bayesian MCMC Methodology: Applies a hierarchical Bayesian Markov Chain Monte Carlo approach to obtain posterior estimates for dissociation constants, sequence-specific binding energies, and free TF concentrations.
• Joint Estimation of Dissociation Constants: Jointly estimates dissociation constants by integrating inferred binding degrees with a model of binding energetics, accounting for data depth and model explanatory power.
• Comprehensive Binding Repertoire Elucidation: Enables elucidation of the full TF-binding repertoire, i.e., the set of sequences bound by TFs with at least moderate strength.

Scientific Applications:

• Transcription Factor Research: Characterizes TF-DNA binding affinities and sequence-dependent binding energies to study transcriptional regulation mechanisms.
• Genomic Studies: Maps potential regulatory elements by identifying sequences bound by specific TFs across genomes.
• Drug Discovery and Development: Provides TF-DNA interaction data that can inform strategies targeting gene-expression regulation.

Methodology:

Uses sequencing counts from pre-bound and bound pools as input to a hierarchical Bayesian MCMC framework that infers posterior distributions for dissociation constants, sequence-specific binding energies, and free TF concentrations, with joint estimation of Kd values based on inferred binding degrees and an explicit binding-energy model that depends on the number of sequence reads and the model's explanatory power.