skip to main content
DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: BraggNet: integrating Bragg peaks using neural networks

Abstract

Neutron crystallography offers enormous potential to complement structures from X-ray crystallography by clarifying the positions of low-Z elements, namely hydrogen. Macromolecular neutron crystallography, however, remains limited, in part owing to the challenge of integrating peak shapes from pulsed-source experiments. To advance existing software, this article demonstrates the use of machine learning to refine peak locations, predict peak shapes and yield more accurate integrated intensities when applied to whole data sets from a protein crystal. The artificial neural network, based on the U-Net architecture commonly used for image segmentation, is trained using about 100 000 simulated training peaks derived from strong peaks. After 100 training epochs (a round of training over the whole data set broken into smaller batches), training converges and achieves a Dice coefficient of around 65%, in contrast to just 15% for negative control data sets. Integrating whole peak sets using the neural network yields improved intensity statistics compared with other integration methods, including k-nearest neighbours. These results demonstrate, for the first time, that neural networks can learn peak shapes and be used to integrate Bragg peaks. In conclusion, it is expected that integration using neural networks can be further developed to increase the quality of neutron, electronmore » and X-ray crystallography data.« less

Authors:
ORCiD logo [1]; ORCiD logo [1]; ORCiD logo [2];  [3];  [4]; ORCiD logo [1]; ORCiD logo [1]; ORCiD logo [1]
  1. Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
  2. University of Nebraska Medical Center, Omaha, NE (United States)
  3. Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States); Princeton University, NJ (United States)
  4. Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States); Lentigen Technologies, Gaithersburg, MD (United States)
Publication Date:
Research Org.:
Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
Sponsoring Org.:
USDOE Office of Science (SC), Basic Energy Sciences (BES); National Institutes of Health (NIH)
OSTI Identifier:
1649618
Grant/Contract Number:  
AC05-00OR22725; R01-GM071939
Resource Type:
Accepted Manuscript
Journal Name:
Journal of Applied Crystallography (Online)
Additional Journal Information:
Journal Name: Journal of Applied Crystallography (Online); Journal Volume: 52; Journal Issue: 4; Journal ID: ISSN 1600-5767
Publisher:
International Union of Crystallography
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICS AND COMPUTING; Integration; machine learning; neural networks; neutron crystallography; computational modelling

Citation Formats

Sullivan, Brendan T., Archibald, Richard, Azadmanesh, Jahaun, Vandavasi, Venu Gopal, Langan, Patricia S., Coates, Leighton, Lynch, Vickie E., and Langan, Paul. BraggNet: integrating Bragg peaks using neural networks. United States: N. p., 2019. Web. https://doi.org/10.1107/s1600576719008665.
Sullivan, Brendan T., Archibald, Richard, Azadmanesh, Jahaun, Vandavasi, Venu Gopal, Langan, Patricia S., Coates, Leighton, Lynch, Vickie E., & Langan, Paul. BraggNet: integrating Bragg peaks using neural networks. United States. https://doi.org/10.1107/s1600576719008665
Sullivan, Brendan T., Archibald, Richard, Azadmanesh, Jahaun, Vandavasi, Venu Gopal, Langan, Patricia S., Coates, Leighton, Lynch, Vickie E., and Langan, Paul. Fri . "BraggNet: integrating Bragg peaks using neural networks". United States. https://doi.org/10.1107/s1600576719008665. https://www.osti.gov/servlets/purl/1649618.
@article{osti_1649618,
title = {BraggNet: integrating Bragg peaks using neural networks},
author = {Sullivan, Brendan T. and Archibald, Richard and Azadmanesh, Jahaun and Vandavasi, Venu Gopal and Langan, Patricia S. and Coates, Leighton and Lynch, Vickie E. and Langan, Paul},
abstractNote = {Neutron crystallography offers enormous potential to complement structures from X-ray crystallography by clarifying the positions of low-Z elements, namely hydrogen. Macromolecular neutron crystallography, however, remains limited, in part owing to the challenge of integrating peak shapes from pulsed-source experiments. To advance existing software, this article demonstrates the use of machine learning to refine peak locations, predict peak shapes and yield more accurate integrated intensities when applied to whole data sets from a protein crystal. The artificial neural network, based on the U-Net architecture commonly used for image segmentation, is trained using about 100 000 simulated training peaks derived from strong peaks. After 100 training epochs (a round of training over the whole data set broken into smaller batches), training converges and achieves a Dice coefficient of around 65%, in contrast to just 15% for negative control data sets. Integrating whole peak sets using the neural network yields improved intensity statistics compared with other integration methods, including k-nearest neighbours. These results demonstrate, for the first time, that neural networks can learn peak shapes and be used to integrate Bragg peaks. In conclusion, it is expected that integration using neural networks can be further developed to increase the quality of neutron, electron and X-ray crystallography data.},
doi = {10.1107/s1600576719008665},
journal = {Journal of Applied Crystallography (Online)},
number = 4,
volume = 52,
place = {United States},
year = {2019},
month = {7}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record

Save / Share:

Works referenced in this record:

Cheetah : software for high-throughput reduction and analysis of serial femtosecond X-ray diffraction data
journal, May 2014

  • Barty, Anton; Kirian, Richard A.; Maia, Filipe R. N. C.
  • Journal of Applied Crystallography, Vol. 47, Issue 3
  • DOI: 10.1107/S1600576714007626

Linking Crystallographic Model and Data Quality
journal, May 2012


Mantid—Data analysis and visualization package for neutron scattering and μ SR experiments
journal, November 2014

  • Arnold, O.; Bilheux, J. C.; Borreguero, J. M.
  • Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, Vol. 764
  • DOI: 10.1016/j.nima.2014.07.029

The IMAGINE instrument: first neutron protein structure and new capabilities for neutron macromolecular crystallography
journal, September 2013

  • Meilleur, Flora; Munshi, Parthapratim; Robertson, Lee
  • Acta Crystallographica Section D Biological Crystallography, Vol. 69, Issue 10
  • DOI: 10.1107/S0907444913019604

The probability distribution of X-ray intensities
journal, October 1949


Integration of neutron time-of-flight single-crystal Bragg peaks in reciprocal space
journal, May 2014

  • Schultz, Arthur J.; Jørgensen, Mads Ry Vogel; Wang, Xiaoping
  • Journal of Applied Crystallography, Vol. 47, Issue 3
  • DOI: 10.1107/S1600576714006372

The recording and analysis of synchrotron X-radiation Laue diffraction photographs
journal, October 1989

  • Helliwell, J. R.; Habash, J.; Cruickshank, D. W. J.
  • Journal of Applied Crystallography, Vol. 22, Issue 5
  • DOI: 10.1107/S0021889889006564

Neutron and X-ray Crystal Structures of a Perdeuterated Enzyme Inhibitor Complex Reveal the Catalytic Proton Network of the Toho-1 β-Lactamase for the Acylation Reaction
journal, December 2012

  • Tomanicek, Stephen J.; Standaert, Robert F.; Weiss, Kevin L.
  • Journal of Biological Chemistry, Vol. 288, Issue 7
  • DOI: 10.1074/jbc.M112.436238

The Cambridge Structural Database
journal, April 2016

  • Groom, Colin R.; Bruno, Ian J.; Lightfoot, Matthew P.
  • Acta Crystallographica Section B Structural Science, Crystal Engineering and Materials, Vol. 72, Issue 2, p. 171-179
  • DOI: 10.1107/S2052520616003954

The finer things in X-ray diffraction data collection
journal, October 1999

  • Pflugrath, J. W.
  • Acta Crystallographica Section D Biological Crystallography, Vol. 55, Issue 10
  • DOI: 10.1107/S090744499900935X

From machine learning to deep learning: progress in machine intelligence for rational drug discovery
journal, November 2017


DIALS : implementation and evaluation of a new integration package
journal, February 2018

  • Winter, Graeme; Waterman, David G.; Parkhurst, James M.
  • Acta Crystallographica Section D Structural Biology, Vol. 74, Issue 2
  • DOI: 10.1107/S2059798317017235

A suite-level review of the neutron single-crystal diffraction instruments at Oak Ridge National Laboratory
journal, September 2018

  • Coates, L.; Cao, H. B.; Chakoumakos, B. C.
  • Review of Scientific Instruments, Vol. 89, Issue 9
  • DOI: 10.1063/1.5030896

Towards automated crystallographic structure refinement with phenix.refine
journal, March 2012

  • Afonine, Pavel V.; Grosse-Kunstleve, Ralf W.; Echols, Nathaniel
  • Acta Crystallographica Section D Biological Crystallography, Vol. 68, Issue 4
  • DOI: 10.1107/S0907444912001308

A review on image segmentation techniques
journal, September 1993


Protein crystallography with spallation neutrons: collecting and processing wavelength-resolved Laue protein data
journal, March 2004


SwissFEL: The Swiss X-ray Free Electron Laser
journal, July 2017

  • Milne, Christopher; Schietinger, Thomas; Aiba, Masamitsu
  • Applied Sciences, Vol. 7, Issue 7
  • DOI: 10.3390/app7070720

LAUEGEN , an X-windows-based program for the processing of Laue diffraction data
journal, April 1995


XDS
journal, January 2010

  • Kabsch, Wolfgang
  • Acta Crystallographica Section D Biological Crystallography, Vol. 66, Issue 2
  • DOI: 10.1107/S0907444909047337

A statistic for local intensity differences: robustness to anisotropy and pseudo-centering and utility for detecting twinning
journal, June 2003

  • Padilla, Jennifer E.; Yeates, Todd O.
  • Acta Crystallographica Section D Biological Crystallography, Vol. 59, Issue 7
  • DOI: 10.1107/S0907444903007947

Classification of crystallization outcomes using deep convolutional neural networks
journal, June 2018


Recent developments in CrystFEL
journal, March 2016

  • White, Thomas A.; Mariani, Valerio; Brehm, Wolfgang
  • Journal of Applied Crystallography, Vol. 49, Issue 2
  • DOI: 10.1107/S1600576716004751

The European X-ray free-electron laser facility in Hamburg
journal, December 2011

  • Altarelli, M.
  • Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms, Vol. 269, Issue 24
  • DOI: 10.1016/j.nimb.2011.04.034

Neutron macromolecular crystallography with LADI-III
journal, October 2010

  • Blakeley, Matthew P.; Teixeira, Susana C. M.; Petit-Haertlein, Isabelle
  • Acta Crystallographica Section D Biological Crystallography, Vol. 66, Issue 11
  • DOI: 10.1107/S0907444910019797

Protein structures by spallation neutron crystallography
journal, April 2008

  • Langan, Paul; Fisher, Zoë; Kovalevsky, Andrii
  • Journal of Synchrotron Radiation, Vol. 15, Issue 3
  • DOI: 10.1107/S0909049508000824

Integration of Bragg reflections with an Anger camera area detector
journal, April 1989


Neutron structure analysis using the IBARAKI biological crystal diffractometer (iBIX) at J-PARC
journal, October 2010

  • Tanaka, Ichiro; Kusaka, Katsuhiro; Hosoya, Takaaki
  • Acta Crystallographica Section D Biological Crystallography, Vol. 66, Issue 11
  • DOI: 10.1107/S0907444910033020

A convolutional neural network-based screening tool for X-ray serial crystallography
journal, April 2018

  • Ke, Tsung-Wei; Brewster, Aaron S.; Yu, Stella X.
  • Journal of Synchrotron Radiation, Vol. 25, Issue 3, p. 655-670
  • DOI: 10.1107/S1600577518004873

HKL -3000: the integration of data reduction and structure solution – from diffraction images to an initial model in minutes
journal, July 2006

  • Minor, Wladek; Cymborowski, Marcin; Otwinowski, Zbyszek
  • Acta Crystallographica Section D Biological Crystallography, Vol. 62, Issue 8
  • DOI: 10.1107/S0907444906019949

Wish: The New Powder and Single Crystal Magnetic Diffractometer on the Second Target Station
journal, April 2011


The Macromolecular Neutron Diffractometer MaNDi at the Spallation Neutron Source
journal, July 2015

  • Coates, Leighton; Cuneo, Matthew J.; Frost, Matthew J.
  • Journal of Applied Crystallography, Vol. 48, Issue 4
  • DOI: 10.1107/S1600576715011243

Processing of X-ray snapshots from crystals in random orientations
journal, July 2014

  • Kabsch, Wolfgang
  • Acta Crystallographica Section D Biological Crystallography, Vol. 70, Issue 8
  • DOI: 10.1107/S1399004714013534

Measures of the Amount of Ecologic Association Between Species
journal, July 1945


Macromolecular Cryocrystallography
journal, June 1997


PHENIX: a comprehensive Python-based system for macromolecular structure solution
journal, January 2010

  • Adams, Paul D.; Afonine, Pavel V.; Bunkóczi, Gábor
  • Acta Crystallographica Section D Biological Crystallography, Vol. 66, Issue 2, p. 213-221
  • DOI: 10.1107/S0907444909052925

Improving the accuracy and resolution of neutron crystallographic data by three-dimensional profile fitting of Bragg peaks in reciprocal space
journal, October 2018

  • Sullivan, Brendan; Archibald, Rick; Langan, Patricia S.
  • Acta Crystallographica Section D Structural Biology, Vol. 74, Issue 11
  • DOI: 10.1107/S2059798318013347

An introduction to data reduction: space-group determination, scaling and intensity statistics
journal, March 2011

  • Evans, Philip R.
  • Acta Crystallographica Section D Biological Crystallography, Vol. 67, Issue 4
  • DOI: 10.1107/S090744491003982X

Status of the neutron time-of-flight single-crystal diffraction data-processing software STARGazer
journal, October 2018

  • Yano, Naomine; Yamada, Taro; Hosoya, Takaaki
  • Acta Crystallographica Section D Structural Biology, Vol. 74, Issue 11
  • DOI: 10.1107/S2059798318012081

The influence of twinning by merohedry on intensity statistics
journal, July 1980

  • Rees, D. C.
  • Acta Crystallographica Section A: Crystal Physics, Diffraction, Theoretical and General Crystallography, Vol. 36, Issue 4
  • DOI: 10.1107/S0567739480001234

Substrate Binding Induces Conformational Changes in a Class A β-lactamase That Prime It for Catalysis
journal, January 2018

  • Langan, Patricia S.; Vandavasi, Venu Gopal; Cooper, Sarah J.
  • ACS Catalysis, Vol. 8, Issue 3
  • DOI: 10.1021/acscatal.7b04114

Image-based crystal detection: a machine-learning approach
journal, November 2008

  • Liu, Roy; Freund, Yoav; Spraggon, Glen
  • Acta Crystallographica Section D Biological Crystallography, Vol. 64, Issue 12
  • DOI: 10.1107/S090744490802982X

The Protein Data Bank
journal, January 2000


Similarity-based machine learning methods for predicting drug–target interactions: a brief review
journal, August 2013

  • Ding, Hao; Takigawa, Ichigaku; Mamitsuka, Hiroshi
  • Briefings in Bioinformatics, Vol. 15, Issue 5
  • DOI: 10.1093/bib/bbt056

Integration of single-crystal reflections using area multidetectors
journal, October 1988

  • Wilkinson, C.; Khamis, H. W.; Stansfield, R. F. D.
  • Journal of Applied Crystallography, Vol. 21, Issue 5
  • DOI: 10.1107/S0021889888005400

Integration of macromolecular diffraction data using radial basis function networks
journal, November 2000

  • Pokrić, Boris; Allinson, Nigel M.; Helliwell, John R.
  • Journal of Synchrotron Radiation, Vol. 7, Issue 6
  • DOI: 10.1107/S0909049500012929

Intensity statistics in twinned crystals with examples from the PDB
journal, December 2005

  • Lebedev, Andrey A.; Vagin, Alexei A.; Murshudov, Garib N.
  • Acta Crystallographica Section D Biological Crystallography, Vol. 62, Issue 1
  • DOI: 10.1107/S0907444905036759

The integration of macromolecular diffraction data
journal, December 2005

  • Leslie, Andrew G. W.
  • Acta Crystallographica Section D Biological Crystallography, Vol. 62, Issue 1
  • DOI: 10.1107/S0907444905039107

Phaser crystallographic software
journal, July 2007

  • McCoy, Airlie J.; Grosse-Kunstleve, Ralf W.; Adams, Paul D.
  • Journal of Applied Crystallography, Vol. 40, Issue 4
  • DOI: 10.1107/S0021889807021206

Neutron Diffraction Studies of a Class A β-Lactamase Toho-1 E166A/R274N/R276N Triple Mutant
journal, March 2010

  • Tomanicek, Stephen J.; Blakeley, Matthew P.; Cooper, Jonathan
  • Journal of Molecular Biology, Vol. 396, Issue 4
  • DOI: 10.1016/j.jmb.2009.12.036

Fully automatic 3D segmentation of the thoracolumbar spinal cord and the vertebral canal from T2-weighted MRI using K-means clustering algorithm
journal, March 2020

  • Sabaghian, Sahar; Dehghani, Hamed; Batouli, Seyed Amir Hossein
  • Spinal Cord, Vol. 58, Issue 7
  • DOI: 10.1038/s41393-020-0429-3

Cavin1 intrinsically disordered domains are essential for fuzzy electrostatic interactions and caveola formation
journal, February 2021


The oligomeric structures of plant cryptochromes
journal, May 2020