Computational Chemistry and Molecular Modeling

1. Computational Chemistry and Molecular Modeling Lecture 1 08.2.12CH418 computational Chemistry KAIST

2. COMPUTATIONAL CHEMISTRY (Use computers to solve chemical problems) Computation | Theory | Modeling Simulation QUANTUM CHEMISTRY (Quantum mechanics + Chemistry) Molecular Modeling (mimic molecular behavior with models) The underlying physical laws necessary for the mathematical theory of a large part of physics and the whole of chemistry are thus completely known, and the difficulty is only that the exact application of these laws leads to equations much too complicated to be soluble. P.A.M. Dirac 1902-1984 Proc. Roy. Soc(London) 123, 714(1929)

3. Molecular modeling starts from structure determination Selection of calculation methods in comp chem Starting geometry from standard geometry, x-ray, etc. Molecule Is bond formation or breaking important? Are many force field parameters missing? Is it smaller than ca. 100 atoms? Are charges of interest? Are there many closely spaced conformers? Is plenty of computer time available? Is the free energy needed? Is solvation important? Molecular mechanics Quantum mechanics Molecular dynamics or Monte Carlo

4. In practice, Computation, Theory, Modeling are interchangeable and can be regarded as simulations • Why do we perform simulations? • What makes simulations possible? • How do we perform simulations? • What types of things/systems do we simulate? • Engineering aspects of simulations This and following slides are from Ponder’s web page (more relevant to biomolecules and TINKER)

5. Classification of computational biologist - all are simulations for this course

7. Why do we perform simulations (especially for biological molecules) ? • Simulations are experiments performed on a computer • (Repeat) Simulations are experiments performed on a computer • Simulations represent a controlled environment • Probe length and time scales not accessible by other means

9. Advantage of simulations • Simulations are not limited by the laws of • physics – we define the laws and basis of • interaction, dynamics, etc. – whether they be • physical or nonphysical • 􀂃 Ignore forces – gravity, viscosity • 􀂃 Use of simple, effective potentials that give the • correct dynamics (simple harmonic oscillator) • 􀂃 Increase time rate – slow reactions, diffusion • limited reactions • 􀂃 Sampling – explore interesting regions of phase • space that may only be periodically visited

10. Connection to conventional wet Lab experiments - Confirm, extrapolate and predict wet lab experimental results – simulations represent just one more experimental tool, just like EM, AFM, Stopped Flow, … • Can be cheaper, faster and easier than wet-lab experiments 􀂃 No protein expression, purification, … 􀂃 Mutations, combinatorial chemistry, etc. is now easy 􀂃 Drug libraries – high throughput screening • Provides details not available in typical experiment 􀂃 Single molecule dynamics – like laser tweezers 􀂃 Reaction pathways 􀂃 Atomic/ Molecular level dynamics – similar to neutron scattering or NMR

11. Ponder ‘ChemBio Lecture note’

12. Disadvantages of simulations - A simulations is only as realistic as you make it – since you control the physics, the physics must be correct • Can be time consuming and approximate • Does not have wide based support from the experimental community (changing slowly) • Some distinguish QM calculations from simulations

13. What makes simulations (of biological systems) possible ? - All sciences are based primarily on numbers, except for biology – this has started to change in the past few decades. • Simulations have become possible since biological information is able to be expressed as data useful for computing (discrete objects) • Protein sequence and structures 􀂃 Sequence information (genome, protein) 􀂃 Fold types 􀂃 Secondary structure elements (alpha helices, beta sheets, …) 􀂃 Tertiary structure (structure of folded protein) 􀂃 Quaternary structures (filaments, polymers)

14. How do we perform simulations? * Simulation does not mean programming! Most simulations techniques, methods, programs, algorithms exist in the public domain. Experimentalists can and do perform good simulations. • Simulations tend to produce lots of data and not just one simple answer. You must understand in detail the physics of a system in order to design a simulation, and you must understand just as much in order to interpret the results. • Simulations can be serial or parallel. Today, most people use PCs for serial jobs and Beowulf clusters or supercomputers for parallel applications

15. What do we simulate? - Historically, simulations were very simple – molecules were spheres, no solvent, … now we can simulate as much detail as we want (if we are willing to pay the price) • Now, we are starting to simulate large biomolecular and nano systems, but challenges still exist • All simulations present a challenge to balance the detail of a simulation with the length of time it takes to run – we can’t afford to simulate everything

16. How do we perform simulations? -This is not theoretical work (most times) since we are not proposing new physical laws or interactions. All we are trying to do is recreate or mimic the physical environment and the interactions between the molecules/proteins involved (remember it’s an experiment) • We use exotic principles such as 􀂃 Newton’s laws – kinematics 􀂃 Coulomb’s law – electrostatics 􀂃 Fluid dynamics 􀂃 Polymer dynamics 􀂃 Schrödinger equation

19. Simulation technology - Increase in computational speed has come at the same time as the biological information • Computer Simulations will be required !! 􀂃 Too time consuming and costly to get structures for all the proteins – the protein folding problem will have to be solved. 􀂃 Once we get protein structures, we need to be able to figure out the physical basis for their interaction in order to understand protein complexes

20. Locating critical points on the potential energy surface(PES) is the major task of molecular simulations • PES: collection of energy values at the given arrangement of atoms • Quantum mechanics • Works with nuclei and electrons • Quantum chemical methods • Classical mechanics (molecular mechanics) • Works with atoms bonded or nonbonded • Usually connection between atoms predefined

21. H/W and S/W issues