Computational networking is all the rage, not only can it help you keep track of your friends and colleagues but it can also reveal a surprising amount of information behind the regulation of gene expression. Now a computationally cutting edge team from the University of Edinburgh have found out how to use transcription factor binding patterns to foresee which histones mod(s) will also be at the scene.
By taking advantage of already known interactions between transcription factors and chromatin modifying enzymes the team was able to show a strikingly accurate prediction of histone modifications based only on previous knowledge of transcription factor binding at promoters and any potential distal regulatory elements. The software was accurate across cell lines and allowed for a large scale analysis. Interestingly, based on the results, the team hypothesizes that “reported associations between histone marks and gene expression may be indirect effects caused by interactions between transcription factors and histone-modifying complexes.”
Senior Author Duncan Sproul shares that “In collaboration with Dr. Guido Sanguinetti, we used machine learning techniques to analyze the relationship between Transcription Factor binding and histone modifications in the human genome. Surprisingly, we show that the distribution of histone modifications in the genome, both at promoters and enhancers, can be accurately predicted from knowledge of the binding patterns of transcription factors. Our results highlight the importance of interactions between transcription factors and the epigenetic machinery in defining epigenetic states in our genomes.”
Check out how much of your research can be done computationally over at PNAS, September 2014