biobb Web Server


BioExcel biomolecular simulation workflows are reaching the production phase, and their power is being demonstrated by technical and scientific success stories. The set of biomolecular simulations building blocks developed for these workflows are already available from github and bioconda. These building blocks are divided in four packages so far:

 

  • biobb_common: base package required to use the biobb
  • biobb_io: module collection to fetch data to be consumed by the rest of the building blocks.
  • biobb_model: module collection to model protein/nucleic acids structures (mutations, missing residues/loops) and run structure quality control (checking).
  • biobb_md: module collection to perform molecular dynamics simulations using GROMACS.

 

 

A new Graphical User Interface (GUI) on top of these workflow building blocks (and eventually on top of all the building blocks developed in BioExcel) is being developed: biobb web server. This GUI will ease the usage of BioExcel workflows and tools for a big community that is still not familiar with HPC programming, but have a real interest on this topic, which can include pharmaceutical companies (used to work with GUIs) but also entry level users, whose interest is demonstrated in the high number of registered users (~4000) and pipelines run as of today with web server interfaces such as MDWeb.

 

            biobb web server will allow users to run a set of chosen, pre-configured workflows built using BioExcel building blocks, such as a structure quality checking, a structure energy minimization, a complete MD setup, or a complete MD simulation (with length restrictions). The GUI will also provide an additional interactivity to our building blocks. A great example is the possibility to run a quality check of a structure, while at the same time a 3D representation of the molecule is shown in the same interface, highlighting the region of the structure of particular interest. This interactivity can be applied also to the set of analyses generated by the workflows.

            Workflows will be submitted and treated by a queue manager, which will serve them in an on-demand processing model performed by Virtual Machines automatically deployed in an Open Nebula OneFlow cloud environment. A direct connection to HPC supercomputing infrastructures to submit long molecular dynamics simulations prepared using the portal will be studied.

 

            biobb server will also be the entry point for BioExcel building blocks. The web page will gather all the information on how to obtain, install and run the building blocks and workflows generated by BioExcel: for developers or experts in the field (github, bioconda, biocontainers, VMs/Cloud) for HPC users (environment modules) and for entry level users (Galaxy, KNIME, web server).

 

 

References:

 

http://mmb.irbbarcelona.org/MDWeb

http://www.charmm-gui.org

https://onlinelibrary.wiley.com/doi/abs/10.1002/jcc.24660