Introduction to PyBMRB 1

Biological Magnetic Resonance data Bank (BMRB)

BMRB 2 is the global archive of NMR spectroscopic data derived from biological molecules like proteins, nucleic acids and metabolites. BMRB collects, annotates, archives, and disseminates (worldwide in the public domain) the important spectral and quantitative data derived from NMR spectroscopic investigations of biologically relevant molecules. The goal is to empower scientists in their analysis of the structure, dynamics, and chemistry of biological systems and to support further development of the field of biomolecular NMR spectroscopy.

NMR-STAR

BMRB uses NMR-STAR 3 data model, which is a data model driven by NMR-STAR dictionary . Each BMRB entry includes the meta data such as the information about the biological sample and solvent, experimental conditions, instrument details, author information and so on along with the observed NMR chemical shift data. NMR chemical shift is the core information available in NMR-STAR formatted files provided by BMRB. These files may also contain additional derived data such as restraints, Residual Dipolar couplings (RDC), J-couplings, Chemical Shfit Aniostrophy(CSA) data and so on. BMRB provides python parser (PyNMRSTAR ) to read, write and edit NMR-STAR files. BMRB data is also available through BMRB-API for machine to machine communication and programmatic access.

Why do we need PyBMRB?

BMRB is mostly used by biologists and biochemists who have very little or no programing experience. NMR spectroscopists may want to view the data as NMR spectra and compare with their spectra measured using their sample. PyBMRB can do this in a single command, avoids the hassle of downloading and parsing the data for visualization. This would greatly benefit the research community by allowing them to quickly and easily compare their data with any BMRB entry and visualizing the BMRB data in different types of 2D spectra.

How does it work?

PyBMRB extracts the assigned chemical shift list from NMR-STAR files and combines them using certain rules defined by the NMR experiment type to generate the peak positions. This peak list is displayed on a 2D plane using interactive data visualization tool plotly . It can also generate chemical shift histograms by fetching the data directly from BMRB through BMRB-API.

1

Kumaran Baskaran, Jonathan R Wedell, Eldon L Ulrich, Jeffery C Hoch, and John L Markley. PyBMRB: Data visualization tool for BioMagResBank. In Meghann Agarwal, Chris Calloway, Dillon Niederhut, and David Shupe, editors, Proceedings of the 20th Python in Science Conference, 59–62. 2021. URL: http://conference.scipy.org/proceedings/scipy2021/pdfs/kumaran_baskaran.pdf.

2

Eldon L. Ulrich, Hideo Akutsu, Jurgen F. Doreleijers, Yoko Harano, Yannis E. Ioannidis, Jundong Lin, Miron Livny, Steve Mading, Dimitri Maziuk, Zachary Miller, Eiichi Nakatani, Christopher F. Schulte, David E. Tolmie, R. Kent Wenger, Hongyang Yao, and John L. Markley. BioMagResBank. Nucleic Acids Research, 36(suppl_1):D402–D408, 11 2007. URL: https://doi.org/10.1093/nar/gkm957, arXiv:https://academic.oup.com/nar/article-pdf/36/suppl\_1/D402/7635401/gkm957.pdf, doi:10.1093/nar/gkm957.

3

Eldon L Ulrich, Kumaran Baskaran, Hesam Dashti, Yannis E Ioannidis, Miron Livny, Pedro R Romero, Dimitri Maziuk, Jonathan R Wedell, Hongyang Yao, Hamid R Eghbalnia, Jeffrey C Hoch, and John L Markley. NMR-STAR: comprehensive ontology for representing, archiving and exchanging data from nuclear magnetic resonance spectroscopic experiments. Journal of Biomolecular NMR, 73(1):5–9, feb 2019. URL: https://doi.org/10.1007/s10858-018-0220-3, doi:10.1007/s10858-018-0220-3.