This exploratory seminar discusses three scientific areas where advances in computational methods could lead to dramatic opportunities for scientific discovery: phylogenomics, proteomics, and metagenomics. Each of these areas presents substantial analytical and computational challenges and needs an interdisciplinary effort. Participants include biologists, computer scientists, mathematicians, and statisticians, convened to consider the opportunities and challenges for these three research areas. The increase in biomolecular sequence data from across the “Tree of Life” creates the opportunity to investigate many biological questions, including the origins of life, how environments impact organisms, and the inference of protein structure and function. Metagenomic data (i.e., sequences obtained from environmental samples) present similar (and perhaps even greater) opportunities, including the discovery of new genes and potentially new species. Indeed, the impact on biology from these multi-species comparisons is expected to be enormous. To make these discoveries requires appropriate analytical methods that can handle datasets containing potentially hundreds of thousands of sequences from different organisms, and often of unknown origin. Yet currently available methods either do not run on these large-scale datasets or do not provide adequate accuracy. Thus, to take advantage of the datasets that are now available and the new datasets that will become available in the coming years using the new sequencing technologies, new methods are needed that reliably provide highly accurate analyses of large heterogeneous sequence datasets.