Acceleration of a Molecular Modelling Code for the Analysis and - - PowerPoint PPT Presentation

▶
acceleration of a molecular modelling code for the
SMART_READER_LITE
LIVE PREVIEW

Acceleration of a Molecular Modelling Code for the Analysis and - - PowerPoint PPT Presentation

Jan 02, 2024 371 likes •691 views

GTC, March 17-20, 2015 Silicon Valley Acceleration of a Molecular Modelling Code for the Analysis and Visualization of Weak Interactions between Molecules A. Roussel, J-C.Boisson, H. Deleau, M. Krajecki and E. Hnon GTC, March 17-20, 2015

slide-1
SLIDE 1

Acceleration of a Molecular Modelling Code for the Analysis and Visualization of Weak Interactions between Molecules

  • A. Roussel, J-C.Boisson, H. Deleau,
  • M. Krajecki and E. Hénon

GTC, March 17-20, 2015 Silicon Valley

slide-2
SLIDE 2

Applied theoretical chemistry

  • ICMR = Experimental laboratory « augmented » by theoretical calculations

Models & Prog.

Kinetics, Thermodynamics Ring Free Energy Molecular Docking

Modeling Activities : ICMR Lab

GTC, March 17-20, 2015 Silicon Valley

slide-3
SLIDE 3
  • CReSTIC = computer science laboratory

Parallel and distributed algorithms High-Performance Computing High-Performance Molecular Modeling

Modeling Activities : CReSTIC Lab

➞ Combinatorial optimisation (genetic algorithm, ant/bee colony) ➞ Parallel algorithms for GPU acceleration URCA = the first CUDA Research Center in France

GTC, March 17-20, 2015 Silicon Valley

slide-4
SLIDE 4

Outline

  • Context: docking and scoring functions
  • Methods: AlgoGen, NCI
  • NCI scoring function on GPU
  • Conclusions and perspectives

4 GTC, March 17-20, 2015 Silicon Valley

slide-5
SLIDE 5

Docking

Ligand

+

5

Macro-molecule

(site)

GTC, March 17-20, 2015 Silicon Valley

slide-6
SLIDE 6

Docking tools

  • Combination of:

– A solution representation

quaternion, torsion, …

– An associated search space according to data flexibility

6 GTC, March 17-20, 2015 Silicon Valley

slide-7
SLIDE 7

Docking tools

– An associated search space according to data flexibility:

  • No flexibility  rigid docking:

– Key / lock paradigm – Basic good interaction information

7 GTC, March 17-20, 2015 Silicon Valley

slide-8
SLIDE 8

Docking tools

– An associated search space according to data flexibility:

  • No flexibility  rigid docking:

– Key / lock paradigm – Basic good interaction information

  • Ligand flexibility  semi-flexible docking:

– Conformation adaptation of the ligand to fit the site

8 GTC, March 17-20, 2015 Silicon Valley

slide-9
SLIDE 9

Docking tools

– An associated search space according to data flexibility:

  • No flexibility  rigid docking:

– Key / lock paradigm – Basic good interaction information

  • Ligand flexibility  semi-flexible docking:

– Conformation adaptation of the ligand to fit the site

  • Ligand and site flexible  full-flexible docking.

– Case of unapproachable site. – Depending of the molecule size: from conformation adaptation of the lateral chains to backbone folding

9 GTC, March 17-20, 2015 Silicon Valley

slide-10
SLIDE 10

Docking tools

– An optimization procedure:

  • Only one method:

– genetic algorithm, ant/bee colony, …

  • cooperative approaches:

– Lamarckian algorithm, …

– A scoring function evaluation of the ligand/site complex quality  energy (main objective)

10 GTC, March 17-20, 2015 Silicon Valley

slide-11
SLIDE 11

Scoring functions

  • Parameterized force field:

– Empirical definition of molecular interactions – Pros:

  • Very fast  only few seconds on big systems
  • Well integrated in tool suite: Autodock, Glide, …
  • Enables full-flexible docking

11 GTC, March 17-20, 2015 Silicon Valley

slide-12
SLIDE 12

Scoring functions

  • Parameterized force field:

– Empirical definition of molecular interactions – Cons:

  • Each molecular family  specific parameters
  • Not able to describe all realistic interactions
  • Substantial input preparation needed

12 GTC, March 17-20, 2015 Silicon Valley

slide-13
SLIDE 13

Scoring functions

  • Quantum mechanics:

– Strict exploitation of electronic information – Pros:

  • No need of (empirical) parameters
  • All the interactions can be described
  • No specific input preparation

13 GTC, March 17-20, 2015 Silicon Valley

slide-14
SLIDE 14

Scoring functions

  • Quantum mechanics:

– Strict exploitation of electronic information – Cons:

  • Very (very) slow: several hours to days for small

systems

  • Not (yet) dedicated for docking analysis:

Rigid docking only

14 GTC, March 17-20, 2015 Silicon Valley

slide-15
SLIDE 15

Outline

  • Context: docking and scoring functions
  • Methods: AlgoGen, NCI
  • NCI scoring function on GPU
  • Conclusions and perspectives

15 GTC, March 17-20, 2015 Silicon Valley

slide-16
SLIDE 16

AlgoGen

  • Framework for rigid quantum docking

based on:

– A genetic algorithm as optimization method – No specific evaluation scoring:

  • Divcon, Mopac, …
  • Gaussian, …

– A master/slave parallel model

16 GTC, March 17-20, 2015 Silicon Valley

slide-17
SLIDE 17

Algogen

2009 Preliminary version AlgoGen-Divcon1

1Thiriot, E.; Monard, G. THEOCHEM. 2009, 898, 31–41. 2Barberot and al., Comp.Theor. Chem. 2014, 1028, 7-18.

2013 Validated version AlgoGen-Mopac2 Barberot C. PhD (ICMR) Thiriot E. PhD (SRSMC) 2014 Validated version AlgoGen-Mopac/NCI 2015 New PhD

17

NCI/GPU/LS

GTC, March 17-20, 2015 Silicon Valley

slide-18
SLIDE 18

NCI

  • New method to predict, visualize and

interprete Non Convalent molecular Interactions

  • Electron density ρ(r)

Electron density gradient ∇ρ(r) Electron density hessien 18

Contreras-Garcia, J. and al, J. Phys. Chem. A. 2011,115, 12983.

GTC, March 17-20, 2015 Silicon Valley

slide-19
SLIDE 19

NCI Post-treatment

PDE4D-zardaverine interactions Zardaverine inhibitor PhosphoDiesterase 4D NCI interaction surfaces

19 GTC, March 17-20, 2015 Silicon Valley

slide-20
SLIDE 20

NCI as a score

  • NCI: based on a grid of atom interactions

describing attraction/repulsion forces

  • Each point can be computed individually
  • Natural parallel scheme:

 from NCI grid to GPU grid

20 GTC, March 17-20, 2015 Silicon Valley

slide-21
SLIDE 21

Outline

  • Context: docking and scoring functions
  • Methods: AlgoGen, NCI
  • NCI scoring function on GPU
  • Conclusions and perspectives

21 GTC, March 17-20, 2015 Silicon Valley

slide-22
SLIDE 22

Methodology

  • Direct use of Fortran code to CUDA
  • Isolation of specific structures and

transformation to one-dimension arrays

  • Thread repartition with redundant calculi

22 GTC, March 17-20, 2015 Silicon Valley

slide-23
SLIDE 23

Input data

  • Test on 3 quantum instances +1 molecular

docking instance (CCDC Astex dataset)

Instance Name Number of atoms in the NCI Grid 3bench2 313 4bench3 326 5bench4 497 6rsa 1666

23 GTC, March 17-20, 2015 Silicon Valley

slide-24
SLIDE 24

Romeo HPC Tesla Cluster

5th 3131 MFLOPS/W

Bull Cool Cabinet Door

151th 254.9 Tflops

Linpack

130 Bull servers

bullx R421 E3 – Bull AE & MPI

VirtualGL technology servers Quadro 6000 & 5800 NVIDIA GRID + Citrix Virtualisation NVIDIA VGX K2 Scalable Graphics 3D cloud solution NVIDIA K6000

260 INTEL Ivy Bridge E5-2650 v2 Processor, non-blocking Mellanox Infiniband, Slurm, 88 To Lustre (NetApp), 57 To home, 100 To Storage 260 NVIDIA Tesla K20X accelerators Big Data, on-demand and remote

Displaying Computing GTC, March 17-20, 2015 Silicon Valley

slide-25
SLIDE 25

GPU Accelerator

  • Nvidia Tesla K20X (Kepler):

– 2688 processor cores – 6 GB GDDR5 – Peak performance:

  • 1.31 Tflops (double-precision floating point)
  • 3.95 Tflops (single-precision floating point)

25 GTC, March 17-20, 2015 Silicon Valley

slide-26
SLIDE 26

Proof of concept results

  • CPU Intel Ivy Bridge (8 cores) vs Tesla

K20X:

– Equivalent purchase and exploitation price

  • Sequential CPU vs :

– OpenMP (8): computation time / 4 – Tesla K20X: computation time / 300

  • OpenMP (8) vs Tesla K20X

– Computation time / 75

26 GTC, March 17-20, 2015 Silicon Valley

slide-27
SLIDE 27

AlgoGen NCI GPU

  • Extrapolated results:

– AlgoGen NCI (on a small system)

  • CPU version  16000 evaluations * 2min

 22 days

  • GPU version  16000 evaluations * 0.4 s

 < 2h

27 GTC, March 17-20, 2015 Silicon Valley

slide-28
SLIDE 28

Outline

  • Context: docking and scoring functions
  • Methods: AlgoGen, NCI
  • NCI scoring function on GPU
  • Conclusions and perspectives

28 GTC, March 17-20, 2015 Silicon Valley

slide-29
SLIDE 29

Conclusions and perspectives

  • The proof of concept is valid
  • Next steps:

– Production phase – Pipeline of evaluations – NCIPLOT code extraction and optimization

29 GTC, March 17-20, 2015 Silicon Valley

slide-30
SLIDE 30

Conclusions and perspectives

  • Application of NCI to docking

– submitted French ANR project by NCI authors (E- NERGY).

  • New PhD:

– New scoring methods

  • Including collaboration with the authors of DFTB codes

(CSC group, Brême, Germany; LCPQ Toulouse, France, LCT group, Paris, France)

– Flexibility management

  • Including collaboration with Marie Brut (LAAS Toulouse)

30 GTC, March 17-20, 2015 Silicon Valley