525 From Sequence Variation to Biomolecular Structure and Function
Alberto Perez, Stony Brook University
The recent explosion of sequence knowledge brings around unprecedented power to use evolutionary tools to understand biomolecular structure. Can we go beyond our current limitations in predicting structure or functionality by combining sequence and structure together?
For protein structures that were beyond experimental techniques and and bioinformatics methods based on known proteins, this yields a promising way forward. Not only for soluble proteins but also for membrane proteins, whose presence in the PDB is still small despite their huge importance as drug targets.
Furthermore, the greater conservation of certain protein regions, like the active site, make the use of these sequence co-evolution methods specially interesting for learning about active site geometry in cases where the whole protein remains challenging.
In the DNA world, it is still an unsolved matter how transcription factors (TF) find promoter regions in order to express genes. TF are known to be promiscuous (binding different DNA sequences). At the same time, each binding site is statistically present millions of times through the genome – and yet most of the binding sites are not bound by TF. Sequence alone is not enough to understand the role of DNA in gene regulation. Recently the combination of structural features from DNA with sequence information is yielding higher predictive power to understand gene regulation.
For protein structures that were beyond experimental techniques and and bioinformatics methods based on known proteins, this yields a promising way forward. Not only for soluble proteins but also for membrane proteins, whose presence in the PDB is still small despite their huge importance as drug targets.
Furthermore, the greater conservation of certain protein regions, like the active site, make the use of these sequence co-evolution methods specially interesting for learning about active site geometry in cases where the whole protein remains challenging.
In the DNA world, it is still an unsolved matter how transcription factors (TF) find promoter regions in order to express genes. TF are known to be promiscuous (binding different DNA sequences). At the same time, each binding site is statistically present millions of times through the genome – and yet most of the binding sites are not bound by TF. Sequence alone is not enough to understand the role of DNA in gene regulation. Recently the combination of structural features from DNA with sequence information is yielding higher predictive power to understand gene regulation.
