WebFF: Force-field repository for organic and soft materials

Summary

The vision of the Materials Genome Initiative (MGI) is a new R&D paradigm enabling accelerated materials development at lower cost via an increased reliance on computation. This new paradigm will be fostered through the creation of national infrastructure for materials data sharing and analysis that supports integrated multiscale modeling. In this hierarchical approach to materials modeling, atomistic molecular dynamics (MD) calculations are integral to calculating properties and phenomena at length scales ranging from Angstroms to tens of nanometers, on time scales from femto- to microseconds. Soft materials modeling is augmented by united atom, coarse-grained and higher level MD models which are based on similar techniques and principles, and enable computation at higher level mesoscale regimes between the atomic and continuum.

A critical data need required to support MD at all hierarchical levels of materials modeling are widely available, curated force-field data, with traceable data provenance. In this project, we are developing an open and extensible force-field repository designed to support at MD calculations at multiple levels of granularity for organic soft materials such as polymers, colloids, gels, composites as well as pharmaceutical and biological materials.

WebFF - A web-hosted, force-field repository for soft materials bulit on the Materials Data Curation System (MDCS).

Description

WebFF is an open and extensible force-field (FF) repository, designed to support the Materials Genome Initiative (MGI) for organic and related soft materials. The repository is built using the NIST Materials Data Curation System (MDCS) which supports ontology based database descriptions using XML schema. Here are some of the main features of WebFF:

• Users interact with the repository through two main portals. The Data Curation Portal supports upload of published force-field data with appropriate metadata descriptors to support provenance based data sharing. New datasets may be curated interactively or using a python based toolset to upload large datasets en masse. Data curation requires an authorized account.
• The Data Exploration Portal supports search for force-field data based on the curated metadata descriptors and download in a number for common formats. We will work with the user community to expand output coverage.
• The initial release of the repository will feature four integrated XML schemas: (1) Class I organic force-fields such as OPLS, Amber and CHARMM style representations; (2) Class II style force-fields such as CFF, PCFF, COMPASS and TEAMFF; (3) Reduced order models such as united atom (UA) and coarse-grained models (CG); (4) A specialized schema for water force-fields such a TIP3P and other related models.
• Documentation website (under construction): https://webff-documentation.readthedocs.io/en/latest/index.html
• Repository website (coming soon): http://webff.nist.gov/