The beauty and the power of protein structures as revealed by ALEPH
What is it about?
Protein structures are not only beautiful but more importantly they are essential to determine what is the role of each protein in the "day-life" of any living cell. We already know a lot about those structures: we have classified them, organised them, we have established relationship among them but still there is much more to understand. Answering questions such as "How a protein fold into a three dimensional structure?", "Why assuming that shape and not any other?", "Can we predict the shape of a structure from just its amino acid sequence?", it is the dream of any structural bioinformatician. With our program ALEPH we want to provide a new way of accessing and interpret the data currently available by describing any protein structure as a network of special vectors derived from atom coordinates. We focus on revealing the detail of very local structural properties to reuse this knowledge in different contexts and determine the structure of unknown proteins.
Why is it important?
We think that with the amount of data available and the outstanding improvements of machine learning algorithms we are now in the perfect moment to explore new ideas and new concepts on how look protein structures. We provide a different interpretation that can be sometime alternative but most of the time complementary to the state of the art software.