Workshop on Molecular Geometry and Visualization

Date: September 11th 2015
Venue: Centro de Congressos do IST, Room 02.2
Programme


9h30

Joaquim Jorge 
(IST-UL)


Opening session

9h45

João Madeiras Pereira
(IST-UL)


Photorealism & Visualisation

10h00

Abel
 Gomes 
(UBI)


Geometric Detection of Pocket Cavities

10h20

Miguel Prazeres
(IST-UL)


Exploring Biomolecular Interactions in Paper-based Biosensors
10h45  Coffee break

11h15

Chandrajit Bajaj 
(UT Austin) 

Predictive and Scalable Macro­Molecular Modeling

12h30  Lunch time

14h30
 
Sérgio Dias (UBI)

Molecular Visualization via Polygonization


15h00

Daniel Simões Lopes (INESC-ID)


The A-MOP Project
 15h30  Coffee break

 16h30
  
Technical discussions + Next steps



Funding:

This workshop is supported by the Fundação para a Ciência e Tecnologia through the project A-MOP - Algorithms for Macro-Molecular Pocket Detection, UTAP-EXPL/QEQ-COM/0019/2014.

UT Austin|Portugal: International Collaboratory for Emerging Technologies, CoLab  

Presentation Abstracts & Bios

Predictive and Scalable Macro­Molecular Modeling

Most bio­molecular complexes involve three or more molecules, forming macro­molecules consisting of thousands to a million atoms. We consider fast molecular modeling algorithms and data structures to support automated prediction of bimolecular structure assemblies formulating
it as the approximate solution of a non­convex geometric optimization problem. The conformation of the macro­molecules with respect to each other are optimized with respect to a hierarchical interface matching score based on molecular energetic potentials ((Lennard­Jones, Coulombic, generalized Born, Poisson Boltzmann ). The assembly prediction decision procedure involves both search and scoring over very high dimensional spaces, (O(6^n) for n rigid molecules), and moreover is provably NP­hard. To make things even more complicated, predicting bio­molecular complexes requires search optimization to include molecular flexibility and induced conformational changes as the assembly interfaces complementarily align. I shall also briefly present fast computation methods which run on commodity multicore CPUs and manycore GPUs. The key idea is to trade off accuracy of pairwise, long­range atomistic energetics for a higher speed of execution. Our CUDA kernel for GPU acceleration uses a cache­friendly, recursive and linear-space octree data structure to handle very large molecular structures with up to several million atoms. Based on this CUDA kernel, we utilize a hybrid method which simultaneously exploits both CPU and GPU cores to provide the best performance based on selected parameters of the approximation scheme.

Chandrajit  Bajaj  is a Professor of Computer Science, and the director of the Center for Computational Visualization in the Institute for Computational and Engineering Sciences (ICES) at the University of Texas at Austin.  Bajaj holds the Computational Applied Mathematics Chair in Visualization. He is also an affiliate faculty member of  Mathematics, Bio-medical Engineering, the Institute of  Cell and Molecular Biology and Neurosciences.  He currently serves on the editorial boards for the International Journal of Computational Geometry and Applications, and the ACM Computing Surveys.  He is a fellow of the American Association for the Advancement of Science (AAAS), the Association of Computing Machinery (ACM), and the Institute of Electrical and Electronic Engineers (IEEE).


Exploring Biomolecular Interactions in Paper-based Biosensors

Miguel Prazeres

Biosensors based on bioactive paper are particularly attractive to perform analytical functions outside laboratory settings in the food, health, industrial and environmental areas. One of the keys for the success of paper-based biosensors is to master the ability to immobilize biomolecules, while adequately preserving activity and stability. 
The use of proteins like Carbohydrate Binding Modules (CBMs) that have a natural affinity to cellulose constitutes a promising strategy in this context. The strategy relies on the fusion of the biosensing molecules (e.g. affinity handles, enzymes, oligonucleotides) with CBMs and on their subsequent immobilization on paper via affinity interactions. The availability of such a platform would open up the possibility of developing paper-based sensors for molecular diagnostics, serological testing and quality control monitoring, among others. In this communication, the role that visualization and molecular simulation tools could play in the development of this platform is discussed in view of experimental results obtained so far.

DMF Prazeres holds a PhD in Chemical Engineering and is currently full professor at Instituto Superior Técnico. His research activities are carried out within iBB-Institute for Bioengineering and Biosciences in the areas of Biomolecular and Bioprocess Engineering. Major topics of study include: Biorecognition and biointerfacing for diagnostics, Molecular design of DNA therapeutics, Microbial cell factories and Purifcation of biomolecules. He is the author of >170 papers.


Geometric Detection of Pocket Cavities

Abel Gomes

Molecular cavities are specific regions on the surface of a biomolecule (e.g., a protein) where another molecule (ligand) may bind. Usually, cavities correspond to  voids, pockets, and depressions of the molecular surfaces. The location of such cavities is important to better understand protein functions, as needed in, for example, structure-based drug design. 
This article introduces a method based on the theory of critical points to detect cavities on the molecular surface. The method, called CriticalFinder, is different from other curvature-based methods found in the literature because it does not evaluate the curvature on, but outside, the surface of the molecule. In fact, what is really evaluated is the curvature of the electron density field (generated by the atoms) outside the molecule. To evaluate the performance of CriticalFinder, a comparative study with other five geometric methods (i.e., LIGSITE, PASS, SURFNET, POCASA, and fpocket) was carried out in order to better evaluate the success 
rate of the CriticalFinder method. In addition, it was also carried out a time performance analysis to both CPU/GPU implementations.

Abel Gomes took a BSc degree (5 years) in electrical engineering and
a MSc in computer graphics at University of Coimbra, Portugal. He also 
obtained a PhD degree in geometric modeling at Brunel University, England. His current research interests include geometric modeling, implicit curves and surfaces, and molecular graphics. He has also served as a member of several international committees and editorial review boards related to computer graphics and computational science. His list of publications includes a book published by Springer-Verlag in 2009. He is a licensed Professional Engineer and a member of the ACM and IEEE. He also served as the Head of Department of Computer Science and Engineering at the University of Beira Interior, Covilhã, Portugal.