NSF Award
1636903
Many well-known diseases can be caused by genetic variants (such as point mutations) that affect important protein features such as enzyme active sites. The scientific community has catalogued millions of genetic variants (in genomic databases) and thousands of protein structures (in the Protein Data Bank). However, these two types of information are not linked, or easily linkable, in a manner that makes it easy to explore the relationships between variants and their structural locations. Integrative research, including genetic variation and protein sequence and 3D structure, has been rare or just focusing on a few proteins individually. In this planning project we will promote and facilitate interactions between experts from these communities with the shared goal of developing methods for integrating these data comprehensively. The project will lay groundwork for precision medicine efforts, and will have a significant impact on research on many species for which exploration of the genetic variation among strains or breeds is important. Furthermore, this project will directly impact education: we currently teach several courses per year in proteomics informatics and systems biology, and we will create and use redistributable teaching modules to help students learn to apply these concepts to their research.<br/><br/>Our proposed methodology will enable researchers to "think beyond linear" when interpreting genetic variation. There is currently a strong tendency for scientists in the genomic and mass-spectrometry proteomics communities to think about genome function in linear terms. However, the functional implications of variants (and post-translational modifications) are strongly influenced by their 3-dimensional location on a protein structure. Due to the lack of readily available tools, this leap from a linear position to a 3-dimensional location is rarely made. Our infrastructure will enable analysis at all scales, from mapping individual variants to a single protein, to mapping millions of variants to all available protein sequences and structures. This will in turn enable the discovery and interpretation of spatial patterns as a function of variant frequencies, affected amino acids, tendency to be post-translationally modified, and location within substructures.
