hide
Free keywords:
Base Sequence; Computational Biology/methods; Gene Regulatory Networks/genetics; Humans; Molecular Sequence Data; Mutation/genetics; Polymorphism, Single Nucleotide/genetics; Protein Binding; Transcription Factors/metabolism
Abstract:
The increasing amount of sequence data provides new opportunities and challenges to derive mechanistic models that can link sequence variations to phenotypic diversity. Here we introduce a new computational framework to suggest possible consequences of sequence variations on regulatory networks. Our method, called sTRAP (strap.molgen.mpg.de), analyses variations in the DNA sequence and predicts quantitative changes to the binding strength of any transcription factor for which there is a binding model. We have tested the method against a set of known associations between SNPs and their regulatory consequences. Our predictions are robust with respect to different parameters and model assumptions. Importantly we set an objective and quantifiable benchmark against which future improvements can be compared. Given the good performance of our method, we developed a publicly available tool that can serve as an important starting point for routine analysis of disease-associated sequence regions.
