David Dahl
David Dahl is Assistant Professor of Statistics and develops Bayesian models for protein structure prediction: the problem of determining the native
three-dimensional structure of proteins and peptides based on information only on information from their amino acid sequence. He works on model-based
clustering procedures, Bayesian nonparametric models, and efficient methods for statistical computing. He has several active collaborations, most recently with Dr. David H. Russell (Professor and Head of the Department of
Chemistry at Texas A&M), Dr. Patricia E. Gamble Klein (Associate Professor of Horticultural Sciences and The Institute for Plant Genomics and Biotechnology), and Dr. Jerry W. Tsai (formerly of the Department of Biochemistry and Biophysics at Texas A&M).
As an example of his recent work, Dr. Dahl has developed statistical methods to more accurately refine the initial template towards the (predicted) native structure. Much refinement work focuses on the bivariate angular distribution at each amino acid position along the protein backbone. Dr. Dahl has developed a Bayesian approach to density estimation for bivariate angular data that uses a Dirichlet process mixture model and a bivariate von Mises distribution. Current methods in the field are based on whole position distributions where data is abundant. Their nonparametric density estimation method, however, provides good density estimates for small datasets, thereby permitting the using half positional distribution in the protein refinement process for increased efficiency and accuracy.
