Cloud Based High Performance Virtual Screening in Novel Drug Discovery

Case Study Team


Id Name Institution (type*) Related WG Role in the team
1  Abdurrahman Olğaç Gazi University & Evias Pharmaceutical R&D, Ltd., TR (A,I)  WG3 Coordinator
2  Steffen Möller Rostock University, DE (A)  WG3 Vice-Coordinator

Team members

Id Name Institution (type*) Related WG
1 Andrea Carotti University of Perugia IT (A) WG3
2 Jean Remy Marchand Novartis, CH (I) WG3
3  Horacio Pérez-Sánchez  Catholic University of Murcia ES (A)  WG3
4  José Pedro Cerón Carrasco  Catholic University of Murcia, ES (A)  WG3
5  Marco Aldinucci  University of Turin, IT (A)  WG2
6  Andrea Bracciali  University of Stirling, UK (A)  WG3
7  Cevdet Aykanat  Bilkent University, TR (A)  WG2
8 Roberto Nuti Evias Pharmaceutical R&D, Ltd., IT (I) WG3
9 Sandra Gesing University of Notre Dame, Indiana, USA (A) WG3,WG4
10 Qian Wang Athlone Institute of Technology, IR (A) WG2
11 Simla Olğaç Gazi University TR (A) WG3
12 Aslı Türe Marmara University, TR (A) WG3

*A-academia, I-industry

Addressed Problem

  • Finding a developable hit molecule or a drug candidate
  • Computational needings
  • Method Validation
  • Complexity of big data interwind with diverse tools of analysis.
  • Security Concerns (Public/Private Cloud)

Drug discovery and development requires the integration of multiple scientific and technological disciplines in chemistry and biology and extensive use of information technology. Virtual screening(VS) is one of the methods used in rational drug design and development studies. It may be applied in early stages of drug discovery pipeline.

The number of modular and scalable computational cloud based web platforms are increasing. Such platforms are being developed to try to help researchers during the drug discovery and development pipeline. They are designed to efficiently perform VS, aimed to identify commercially available lead-like and drug-like compounds to be acquired and tested. Chemical datasets can be built, library analysis and profiling, receptor preparation can be done to be used for a molecular docking based VS using cloud technologies. Such platforms could also be adapted to be included in different stages of the R&D process to rationalize the needs.

Existing Solution(-s) (Models, Tools)

All fragments of complete workflow / use case / solutions

Cloud Based

  • Evias Cloud

  • Achilles Blind Docking Server

  • Scripps

  • mCule …


Commercial Databases

  • Zinc

  • Molport

  • eMolecules …


Chemical Vendors

  • Enamine

  • LifeChemicals …



. Schrödinger

. AutoDock …



  • AWS


  • OpenStack

    • Public or Private Clouds

  • HPC Servers

 Proposed Solution(-s) (Models, Tools)

Development of a Rule Based System is required. This system may include an architecture, as described below:

USER: describes the needings of the simulation and the state of analysis
WORKFLOW: describes steps taken (or to be taken) in the simulation( and/or analysis)
SYSTEM: describes rules/constrains/priorisations to get from a well defined current state into another state that is yet unknown but somehow promising, since this is where the reserach with a particular tool has been performed. There will be different rules to select the compounds to screen.
EX-SYSTEM: Information that is contributed from the literature or in vitro experiments that better describe the current state of the analysis.

Practical Scenarios (-s)

    • Blockchaining
    • Similarity based screening
    • Common Workflow Language
    • Description of Key Tools
    • GLUE Language
  • Middleware

Supplementary Material

Useful links

  1. Evias Cloud
  2. Achilles Blind Docking Server
  3. The 22nd EuroQSAR Symposium – Translational and Health Informatics: Implications for Drug Discovery