La Trobe University > LIMS > Research programs > Digital Biology > David Winkler group

David Winkler group

Molecular AI and machine learning

Computation is the third arm of research, after theory and experiment. Computation and simulation of molecular systems is becoming indispensable for 21st century science. However, the size, scale and complexity of realistic materials-biology interactions precludes the application of rigorous, physics-based computational methods like molecular dynamics and quantum chemistry. AI and machine learning are making spectacular inroads into solving some of these very complex problems. Some of our project also involve the use of complex systems science thinking.

We use a wide range of computational chemistry and AI-based methods (principally machine learning) to model complex systems to predict their properties and gain insight into mechanism of interaction at the molecular level. As these are broadly applicable platform methods, we collaborate with experimental scientists across a wide range of projects, in several countries, some of which address non-biological structure-property relationships problems in materials.

Research Areas

We have applied advanced informatics and machine learning methods to extract new knowledge from surface analysis methods such as mass spectra. We are applying these methods to materials and tissue profiling of tumour samples (Gardner et al. Anal Chem 2020) and to libraries of biomaterials that are candidates for coatings for implantable and indwelling medical devices (Burroughs et al. PNAS, 2020). We are also using them to develop new, environmentally benign corrosion inhibitors. Corrosion represent and >US$1 trillion impost on industry.

Working with the University of Nottingham on several large EPSRC projects, we are discovering and designing new materials for medical applications. We use data from high throughput experimentation to build predictive models of the biological effects of biomaterials. We have capabilities to explore a large range of surface chemistries and microtopographies and to capture them as novel mathematical descriptors for training machine learning models (Vallieres et al. Science Adv. 2020; Celiz et al. Nature Mater. 2014).

Climate change, increasing mobility of the world's population, reduced habitats, and ageing populations are all contributing the the spread of existing and often poorly treated neglected tropical diseases. These afflict a very large number of people, largely from developing counties in Africa and South America. the COVID19 pandemic and earlier epidemic caused by coronaviruses show how quickly these can massively impact health and economies on a global scale. This research used Ai and machine learning, together with other computational methods like molecular dynamics and docking to repurpose existing drugs, or find new molecular chemotypes as drug leads. Some of this work can be extended to problems in agricultures, specifically insect control.

AI and machine learning developments are appearing at a phenomenal pace, providing solutions to many problems in science, technology, and medicine that have been inaccessible previously. This research area is developing novel methods for mathematical featurisation of large, complex molecules and biologically-relevant entities like proteins, polymers, nanoparticles,, and micro- and nano topographies. When these are coupled to new AI and machine learning methods they will provide greater accuracy and explainability of AI models developed. We are also using fuzzy systems and information theory to improve featurisation and explainability of models, and are developing open access machine learning tools specifically designed for experimental scientists with little experience with AI and machine learning to model their data.

Meet the Team

Group leader

  • Professor David Winkler

Group PhD researcher

  • Eduardo Aguilar Bejarano (University of Nottingham)

La Trobe University collaborators

  • Professor Paul Pigram
  • Dr Wil Gardner
  • Dr Sarah Bamford

Academic collaborators - Australia

  • Prof Ivan Cole (Australian National University)
  • Dr Nas Mefati (Bio21, University of Melbourne)
  • Prof Tony Hughes (CSIRO)
  • Prof Nico Voelcker (Monash University)
  • Dr Tu Le (RMIT)
  • Prof. Rachel Caruso (RMIT)

Academic collaborators - International

  • Prof Morgan Alexander (University of Nottingham)
  • Dr Grazziela Figieredo (University of Nottingham)
  • Prof Ricky Wildman (University of Nottingham)
  • Dr. Daniel Keddie (University of Nottingham)
  • Prof. Alex Tropsha (UNC Chapel Hill)
  • Prof Arjan Mol (Delft)
  • Prof. Sviatlana Lamaka (Herzberg Institute Hamburg)

Industry collaborators

  • Prof Nikolai Petrovsky (Vaxine Inc.)
  • Dr Sakhi Piplani (Vaxine Inc.)

Patents

  1. Compositions and implantable devices. United States Patent Application No. 17/59575023 November 2021.\

  2. WO2020/237280A1 published patent Composition and implantable devices (antifibrotics), published 03/12/2020.
  3. TW9568/AU/PRV01; Provisional patent application - anti-fibrotic compositions and devices comprising same - draft specification filed 26 May 2019
  4. Compositions for the treatment of fibrosis, provisional patent 18 March 2015. Covers terphenyl compounds and their use in the prevention or treatment of fibrosis. Drug lead A32
  5. Compositions for the treatment of kidney disease, provisional patent 18 March 2015. Covers terphenyl compounds and their use in the prevention or treatment of kidney disease – P5 and related compounds under development in the VIP Compound Library
  6. Radioprotector compounds and methods, publication number US 8999993 B2, application number US 13/640,188, and publication date April 7 2015. Granted and published as USRE46943E1 on 10 July 2018.
  7. Peptidyl TPOR antagonists and uses thereof, US patent application US2018/0193409 A1 12 July 2018
  8. Small molecule TPOR antagonists and uses thereof, PCT filing July 2016 TW9056/WO
  9. Small molecule TPOR antagonists and uses thereof, Australian provisional patent 2015, AU provisional application no. 2015902665
  10. Compositions for the treatment of hypertension and/or fibrosis, WO 2015/039173 A1 published 26 March 2015.
  11. Compositions for the treatment of hypertension and/or fibrosis, WO 2015/039172 A1 published 26 March 2015.
  12. Peptidyl TPOR antagonists and uses thereof, 2013, Australian Provisional Application No. 2013902363.
  13. Radioprotection Compounds and related methods, 35 countries including US, China, India, Korea PCT/AU2011/000392 and WO 2011/123890 A1. US 2013/0109678 A1 published 2 May 2013
  14. Radioprotection Compounds and related methods, US Provisional Patent Application No. 61/321288
  15. Cyclic-Linear Peptidyl TPOR Agonists, Australian Patent Office No: 2008903359, 30th June 2008.
  16. Peptidyl-Small Molecule TPOR Agonists, Australian Patent Office No: 2008903350, 30th June 2008.
  17. Immobilized c-Kit Agonists, Australian Patent Office No: 2008903356, 30th June 2008.
  18. c-Kit Agonists, Australian Patent Office No: 2008903351, 30th June 2008.
  19. Cyclic Peptidyl Dimers, Australian Patent Office No: 2008903335, 30th June 2008.
  20. Immobilized Peptidyl TPOR Agonists, Australian Patent Office No: 2008903333, 30th June 2008.
  21. TPOR Receptor Agonists, Australian Patent Office No: 2008903337, 30th June 2008.
  22. Novel Cyclic Peptidyl Dimers for Therapy, Australian Patent Office No: 2008903334, 30th June 2008.
  23. Novel TPOR Agonists, Australian Patent Office No: 2008903332, 30th June 2008.
  24. Promoters of Thrombopoiesis and Megakaryocytopoiesis, Australian Patent Office No: 2008903330, 30th June 2008.
  25. TPOR Receptor Agonists, Australian Patent Office No: 2008903336, 30th June 2008
  26. Crystalline composition for assessing interaction between compound and BtEcR/BtUSP heterodimer LBD, has BtEcR/BtUSP heterodimer ligand binding domain of ecdysone receptor from Bemisia tabaci, Lawrence M.C., Pilling P., Lovrecz G.O., Epa V.C., Carmichael J.A., Noyce L., Graham L., Hannan G.N., Winkler D., Hill R.J., Patent # WO2004106374-A1
  27. “Compound Screening System” Patent PCT/AU98/00715 filed September 1998. WO 1999012118 A1, EP 1010094 A1
  28. “Compound Screening System” Provisional patent TW 6073/AU filed September 1997.
  29. “Gun Flash Suppressants”, Australian provisional patent PG 7060, 11th September 1984. PCT International Application No. PCT/AU85/00207 filed 30th August 1985 and amended on 28th January 1986. WO 1986001796 A1
  30. “Macrogranular Gun Propellant Charge”, Australian provisional patent PG 6455, 8th August 1984. Full patent PCT/AU85/00208 filed.

Publications

See a full list of publications at: