Sections

• Research
• Teaching
• Toolkit
• BaMBA IV

• ecs20
• ecs229
• FRS03
• ecs 129
• ecs89l
• ecs15
• Documentation

ecs 129 Forewords Introduction Nucleic Acids Proteins: Amino Acids Proteins: Secondary and Tertiary Structures Sequence Comparison Protein Structure Classification RNA Structure Prediction Protein Structure Prediction Lecture Notes Summary Supporting Papers Quiz Protein Protein Interaction Docking Databases Review Quick links pakoehl@ucdavis.edu

Patrice Koehl's teaching web site > ecs 129 > Protein Structure Prediction > Summary Personal tools Log in Summary Protein Structure Prediction: Summary Proteins are most often described using classical mechanics. They are described as a collection of interacting atoms represented by balls. Interactions between atoms can be divided into 2 groups: bonded interactions (bonds, angles and torsion angles), and non bonded interactions (vdW interactions and electrostatics interactions). Structural genomics is based on the premise that homology modeling can generated accurate models of protein structure that are useful to the biologists. Homology modeling proceeds in 4 steps: fold recognition (identification of an homologous protein whose structure is known and serves as a template to build the structure of the unknown protein structure), loop building (filling in the gaps in the template, that correspond to regions in the alignment of the template and target sequences with low similarity, or that contain insertions / deletions), sidechain conformation prediction, and refinement. There are two main types of ab initio protein structure prediction methods: those based on physics that try to mimic the floding of a protein, and those that generate many possible model structures ("decoys"), and then attempt to select those that are native like. CASP is a scheme that was designed to provide conditions to test prediction methods on blind experiments. CASP is biannul, and 2006 was the year of CASP7.

---