Ab Initio Protein Structure Prediction
44 likes35,321 views
Ab initio protein structure prediction uses computational methods to predict a protein's 3D structure from its amino acid sequence. It relies on conformational searching to generate structure decoys and selecting native-like models. The key factors for success are an accurate energy function, efficient search methods like molecular dynamics or genetic algorithms, and effective selection of models close to the native structure. Model selection approaches include energy evaluations, compatibility scores, clustering of similar decoys, and identifying the lowest energy conformations.
Ab Initio Protein Structure Prediction
- 1. Ab Initio Protein Structure Prediction ag1805xag1805x
- 2. Protein structure prediction ● Protein structure prediction (PSP) is the prediction of the three-dimensional structure of a protein from its amino acid sequence i.e. the prediction of its tertiary structure from its primary structure.
- 3. Protein Structure Prediction: Methods Similar Protein Structure Available Not Available Template Based Method ab initio modelling Threading
- 4. ab initio modelling ● ab initio modelling conducts a conformational search under the guidance of a designed energy function. ● This procedure usually generates a number of possible conformations (structure decoys), and final models are selected from them.
- 5. ● A successful ab initio modelling depends on three factors: ➢ an accurate energy function with which the native structure of a protein corresponds to the most thermodynamically stable state, compared to all possible decoy structures; ➢ an efficient search method which can quickly identify the low-energy states through conformational search; ➢ selection of native-like models from a pool of decoy structures.
- 6. Energy Functions ● Energy classified into two groups: ➔ Physics-based energy functions ➔ Knowledge-based energy functions
- 7. Physics-Based Energy Functions “In a strictly-defined physics-based ab initio method, interactions between atoms should be based on quantum mechanics and the coulomb potential with only a few fundamental parameters such as the electron charge and the Planck constant; all atoms should be described by their atom types where only the number of electrons is relevant.” (Hagler et al. 1974; Weiner et al. 1984)
- 8. Physics-Based Energy Functions “In a strictly-defined physics-based ab initio method, interactions between atoms should be based on quantum mechanics and the coulomb potential with only a few fundamental parameters such as the electron charge and the Planck constant; all atoms should be described by their atom types where only the number of electrons is relevant.” (Hagler et al. 1974; Weiner et al. 1984)
- 9. A compromised force field with a large number of selected atom types is used. In each atom type, the chemical and physical properties of the atoms are enough alike with the parameters calculated from crystal packing or quantum mechanical theory.
- 10. ● Well-known examples of such all-atom physics- based force fields include: ✔ AMBER ✔ CHARMM ✔ OPLS ✔ GROMOS96 ● These potentials contain terms associated with bond lengths, angles, torsion angles, van der Waals, and electrostatics interactions. ● The major difference between them lies in the selection of atom types and the interaction parameters.
- 11. Knowledge-Based Energy Function ● Refers to the empirical energy terms derived from the statistics of the solved structures in deposited PDB. ● Can be divided into two types: ➢ generic and sequence-independent terms such as the hydrogen bonding and the local backbone stiffness of a polypeptide chain ➢ amino-acid or protein-sequence dependent terms, e.g. pair wise residue contact potential, distance dependent atomic contact potential , and secondary structure propensities
- 12. Conformational Search Methods ● Successful ab initio modelling of protein structures depends on the availability of a powerful conformation search method which can efficiently find the global minimum energy structure for a given energy function with complicated energy landscape. ● Types: ➔ Monte Carlo Simulations ➔ Molecular Dynamics ➔ Genetic Algorithm ➔ Mathematical Optimization
- 13. Monte Carlo Simulations ● Its core idea is to use random samples of parameters or inputs to explore the behavior of a complex system or process.
- 14. Initial configuration of particles in a system Monte Carlo move is attempted that changes the configuration of the particles Move is accepted or rejected based on an acceptance criterion Calculates the value of a property of interest An accurate average value of this property can be obtained StepsinMCsimulation
- 15. Molecular Dynamics ● MD simulation solves Newton’s equations of motion at each step of atom movement, which is probably the most faithful method depicting atomistically what is occurring in proteins. ● The method is therefore most-often used for the study of protein folding pathways ● The long simulation time is one of the major issues of this method, since the incremental time scale is usually in the order of femtoseconds (10 15 s) while the fastest folding time of a small− protein (less than 100 residues) is in the millisecond range in nature.
- 16. Genetic Algorithm ● The genetic algorithm is a method for solving problems that is based on natural selection, the process that drives biological evolution. ● The genetic algorithm repeatedly modifies a population of individual solutions. ● At each step, the genetic algorithm selects individuals at random from the current population to be parents and uses them to produce the children for the next generation. ● Over successive generations, the population "evolves" toward an optimal solution.
- 17. Mathematical Optimization ● Mathematical optimization is the selection of a best element (with regard to some criteria) from some set of available alternatives.
- 18. Model Selection ● The selection of protein models has been emerged as a new field called Model Quality Assessment Programs (MQAP) ● Modelling selection approaches can be classified into two types: energy based free-energy based
- 19. Physics-Based Energy Function ● Selects the decoy with the lowest energy.
- 20. Knowledge-Based Energy Function ● Sippl developed a pair wise residue-distance based potential (Sippl 1990) using the statistics of known PDB structures in 1990 (its newest version is PROSA II (Sippl 1993; Wiederstein and Sippl 2007) ). ● A variety of knowledge-based potentials have been proposed, which include atomic interaction potential, solvation potential, hydrogen bond potential, torsion angle potential, etc.
- 21. Sequence-Structure Compatibility Function ● Best models are selected not purely based on energy functions. ● They are selected based on the compatibility of target sequences to model structures. ● The earliest and still successful example is that by Luthy et al. (1992), who used threading scores to evaluate structures. ● Colovos and Yeates (1993) later used a quadratic error function to describe the non-covalently bonded interactions among CC, CN, CO, NN, NO and OO, where near-native structures have fewer errors than other decoys
- 22. Clustering of Decoy Structures ● Cluster analysis or clustering is the task of grouping a set of objects in such a way that objects in the same group (called a cluster) are more similar (in some sense or another) to each other than to those in other groups (clusters). ● The cluster-centre conformation of the largest cluster is considered closer to native structures than the majority of decoys. ● In the work by Shortle et al. (1998), for all 12 cases tested, the cluster-centre conformation of the largest cluster was closer to native structures than the majority of decoys. Cluster-centre structures were ranked as the top 1–5% closest to their native structures.
- 24. Fig.: Flowchart of the ROSETTA protocol
- 25. Fig.:Flowchart of I-TASSER protein structure modelling