One of the greatest challenges for computational geometry is to contribute to the understanding of nature in terms of geometry and algorithms through the connection offered by molecular biology. The most difficult of the specific problems mentioned above is the simulation of the folding process that would allow the determination of protein structure from the amino acid residue sequence, and a solution may be out of reach at the moment. The other problems seem tractable, and solutions already exist for easy cases and also for some of the hard ones. The only certainty is that a considerable amount of research is still called for.
Computational geometry can help in several ways. The first is to participate in the design of geometric models. The sphere model for atoms is the fundamental idea connecting particle physics with geometry. There are still open questions as to what extent this simplification is sufficiently accurate and how physical questions can be approached through studying large conglomerates of spheres. Based on various extensions of the spherical model [126][91], the biology community has developed its own geometric software [38]. The geometry is enhanced by graphics and numerical software visualizing and utilizing the geometric information. New software will have to compete with the available packages, which are already widely used, and it will need to follow data and calling standards so components can be plugged together to solve large problems.