Merging NMR Data and Computation Facilitates Data-Centered Research.

Michael R Gryk,Jeffrey C Hoch,Jonathan R Wedell,D Levi Craft,Mark W Maciejewski,Kumaran Baskaran,Hamid R Eghbalnia,Adam D Schuyler,Colin W Wilburn

doi:10.3389/fmolb.2021.817175

Michael R Gryk, Jeffrey C Hoch + Show 7 more

Open Access

https://doi.org/10.3389/fmolb.2021.817175

Copy DOI

Abstract

The Biological Magnetic Resonance Data Bank (BMRB) has served the NMR structural biology community for 40 years, and has been instrumental in the development of many widely-used tools. It fosters the reuse of data resources in structural biology by embodying the FAIR data principles (Findable, Accessible, Inter-operable, and Re-usable). NMRbox is less than a decade old, but complements BMRB by providing NMR software and high-performance computing resources, facilitating the reuse of software resources. BMRB and NMRbox both facilitate reproducible research. NMRbox also fosters the development and deployment of complex meta-software. Combining BMRB and NMRbox helps speed and simplify workflows that utilize BMRB, and enables facile federation of BMRB with other data repositories. Utilization of BMRB and NMRbox in tandem will enable additional advances, such as machine learning, that are poised to become increasingly powerful.

Highlights

AND OVERVIEWThe Biological Magnetic Resonance Data Bank (Ulrich et al, 2008), BMRB, was conceived and launched by John Markley and Eldon Ulrich 40 years ago and has grown to become an indispensable resource for nuclear magnetic resonance (NMR) structural biology, and is emerging as a resource for all aspects of biomolecular NMR, including metabolomics
A key feature of NMRbox is that it provides software persistence by capturing software in virtual machines (VMs) and maintaining an archive of versioned VMs
To aid in software discovery on the NMRbox platform we provide a software registry on the NMRbox website with the ability to search by category

Summary

Introduction

AND OVERVIEWThe Biological Magnetic Resonance Data Bank (Ulrich et al, 2008), BMRB, (bmrb.io) was conceived and launched by John Markley and Eldon Ulrich 40 years ago and has grown to become an indispensable resource for nuclear magnetic resonance (NMR) structural biology, and is emerging as a resource for all aspects of biomolecular NMR, including metabolomics. The chemical shift archive has been impactful for NMR structural biology, contributing to the development of tools such as TALOS (Shen et al, 2009; Shen and Bax, 2015), SPARTA (Shen and Bax, 2007; Shen and Bax, 2010), SHIFTX (Neal et al, 2003; Han et al, 2011), Chemical Shift Index (CSI) (Wishart and Sykes, 1994; Hafsa and Wishart, 2014; Hafsa et al, 2015), and CS-Rosetta (Shen et al, 2008) These tools relied on broad trends originally discovered in the 1960s (Markley et al, 1967) along with their structural correlates revealed by the chemical shift archive and the federated Protein Data Bank (Burley et al, 2017; Burley et al, 2019) (PDB) archive.

Objectives

Methods

Discussion

Conclusion