Bogdan Toader

Postdoctoral Scientist, MRC Laboratory of Molecular Biology

Research interests:

• computational imaging and image processing for cryo-EM and cryo-ET
• machine learning for scientific discovery, aka “AI 4 science”
• computer vision, inverse problems, optimisation, numerical analysis

Email: [email protected]
  Curriculum Vitae: May 2024

Google Scholar | GitHub | ORCID | LinkedIn | Bluesky

About

I am a postdoctoral scientist at the MRC Laboratory of Molecular Biology, developing computational methods for cryo-ET in the groups of Dr Sjors Scheres and Dr Tanmay Bharat.

Previously:

• Postdoc in the Department of Statistics and Data Science and the Quantitative Biology Institute, Yale University: algorithms for cryo-EM
• Postdoc in the Cambridge Advanced Imaging Centre, University of Cambridge: deconvolution for light-sheet microscopy data
• DPhil in the Numerical Analysis Group and InFoMM, University of Oxford, under the supervision of Prof. Jared Tanner and Dr. Andrew Thompson, in collaboration with the National Physical Laboratory: thesis on grid-free compressed sensing/super-resolution
• Core participant of the IPAM long program on computational microscopy at UCLA and an enrichment student at the Alan Turing Institute
• BSc Computer Science and Mathematics, University of Manchester

Research

For a complete list of publications, see Google Scholar.

Selected publications:

• An image processing pipeline for electron cryo-tomography in RELION-5
  A. Burt, B. Toader, R. Warshamanage, A. von Kügelgen, E. Pyle, J. Zivanov, D. Kimanius, T.A.M. Bharat, S.H.W. Scheres
  FEBS Open Bio, 2024


• Efficient high-resolution refinement in cryo-EM with stochastic gradient descent
  B. Toader, M. A. Brubaker, R. R. Lederman
  arXiv:2311.16100, 2023


• On Manifold Learning in Plato's Cave: Remarks on Manifold Learning and Physical Phenomena
  R. R. Lederman, B. Toader
  International Conference on Sampling Theory and Applications (SampTA), 2023


• Methods for Cryo-EM Single Particle Reconstruction of Macromolecules Having Continuous Heterogeneity
  B. Toader, F. J. Sigworth, R. R. Lederman
  Journal of Molecular Biology, 2023


• Image reconstruction in light-sheet microscopy: spatially varying deconvolution and mixed noise
  B. Toader, J. Boulanger, Y. Korolev, M. O. Lenz, J. Manton, C.-B. Schönlieb, L. Mureşan
  Journal of Mathematical Imaging and Vision, 2022


• Sparse non-negative super-resolution --- simplified and stabilised
  A. Eftekhari, J. Tanner, A. Thompson, B. Toader and H. Tyagi
  Applied and Computational Harmonic Analysis, 2021


• The dual approach to non-negative super-resolution: impact on primal reconstruction accuracy
  S. Chrétien, A. Thompson, B. Toader
  2019 13th International Conference on Sampling Theory and Applications (SampTA), 2019

Industry engagement

• Data Study Group, Alan Turing Institute UK, Dec 2019
  Implemented deep learning approach to image segmentation of bright-field microscopy data, problem proposed by the Dstl.
  Report: Dstl – Bright-field image segmentation, https://doi.org/10.5281/zenodo.4452761

• Mathematics in Industry New Zealand Workshop New Zealand, Jun 2019
  Applied optimal transport solution to decompose the spectrum of cheese samples into different components (fat, protein etc.).
  

• 142nd European Study Group with Industry Lithuania, Jun 2018
  Worked on predicting the sustainable income of loan applicants according to central bank rules.
  Report: Sustainable income algorithm, https://doi.org/10.33774/miir-2021-5wvbh

• Data Study Group, Alan Turing Institute UK, Dec 2016
  Implemented network model for the analysis and prediction of air traffic, problem proposed by Airbus.
  Scientific article: Global air transport complex network: multi-scale analysis, https://doi.org/10.1007/s42452-019-0702-2

• InFoMM Mini-Project, Culham Centre for Fusion Energy UK, Jul-Sep 2016
  Investigated numerical methods for computing multiple periodic solutions to a system of ODEs describing the behaviour of plasma.
  Report: Deflating Magnetic Oscillations

• InFoMM Mini-Project, National Physical Laboratory UK, May-Jul 2016
  Analysed how compressed sensing can be applied to the problem of ship localisation from measurements of the sound in the shipping lane.
  Report: Source Reconstruction from Hydrophone Data

• 116th European Study Group with Industry UK, May 2016
  Implemented mixed integer programming solution to the problem of scheduling field trials proposed by Syngenta.