Blog

A Neural Network To Predict How Proteins Change Shape

Proteins shape is constantly changing in response to a variety of biological processes. For example, a protein may be folded into a compact shape and then open up to grab a molecule such as a drug. Predicting what change occurs when you observe proteins behave in an experiment can be difficult. This is because intermediates between two states are by their nature transient, and therefore difficult to observe.

In collaboration with Chris Willcocks in Durham University, We have trained a neural network on protein shapes generated by molecular dynamics simulations, and then asked it to predict possible protein intermediate conformations. We demonstrated that the system can generate conformations that are both physically plausible and close enough to known intermediates. The convolutional architecture of our neural network can accommodate any protein, without the need of being customised. One added advantage that comes with this architecture is transfer learning, which can assist with training on a sparse dataset by first exposing the neural network to a related richer dataset.

This blog post was written with the help of a neural network, that also decided on the title.

V.K. Ramaswamy, S.C. Musson, C.G. Willcocks, M.T. Degiacomi. Learning protein conformational space with convolutions and latent interpolations. Physical Review X 11

Integral membrane protein docking

JabberDock, our protein-protein docking algorithm, has been extended to handle integral membrane proteins. We have created a benchmark set featuring 20 diverse dimers, and used it to demonstrate that our method yields an acceptable solution in its top 10 candidates in 75% of cases.

The first application of JabberDock for membrane proteins has been in the scope of a collaboration with the Kukura group at the University of Oxford. Our models of the protein bo3 oxidase have contributed demonstrating that mass photometry, a novel label-free single molecule technique to measure the mass of biomolecules, can interrogate membrane proteins.

Ray tracing + protein docking = protein tracing

Protein docking essentially involves taking the atomic structure of two proteins, generating a collection of possible molecular arrangements, and scoring their quality to determine the best one. However, the vast majority of arrangements will feature proteins that either overlap of do not touch. As these cases are uninteresting, generating them and assessing their quality is a waste of time.

Enter shape tracing. In collaboration with the Chris Willcocks group in Durham’s Computer Science Department (in particular, shout-out to Adam Leach), we combine our protein volumetric representation (STID maps) with a generalisation of ray racing capable of detecting the collision of arbitrary three-dimensional shapes.

This enables us to very quickly generate arrangements of proteins that are in contact but do not overlap, speeding up the search for the most suitable model.

A. Leach, L.S.P. Rudden, S. Bond-Taylor, J.C. Brigham, M.T. Degiacomi, C.G. Willcocks (2020). Shape tracing: An extension of sphere tracing for 3D non-convex collision in protein docking, 2020 IEEE 20th International Conference on Bioinformatics and Bioengineering (BIBE)

Welcome Lucy!

Lucy, Physics student in Durham, joins us for her Masters project. She will work on using molecular simulations and convolutional neural networks to sample protein conformational spaces.

Mass spectrometry, structural biology and computational methods: 3 articles!

This week we celebrate the publication of three articles in Analytical Chemistry!

MS_review_pic1
How well do we understand different aspects of native mass spectrometry? And their relationships?

In the scope of the EU COST Action BM1403 on Native MS and Related Methods for Structural Biology, we have contributed to the redaction of two articles. These review existing methods and software to analyse and interpret native mass spectrometry data, and  highlight state of the art and current challenges in the area. We hope these works will stimulate discussions within the community, and serve as help for newcomers to the area

MS_erik_michaelIn related news, in collaboration with Michael Landreh (Karolinska Instituet) and Erik Marklund (Uppsala University) we have figured out that ion mobility can tell more than one would think about the shape of protein complexes. Coupling collision cross section and mass measurements with protein databank searches, we have shown that it is possible to determine whether a protein complex is prolate (like a cigar) or oblate (like a disk).

M. Landreh et al. (2020). Predicting the shapes of protein complexes through collision cross section measurements and database searches, Analytical Chemistry

Welcome back Sam!

Sam joins our group again for his PhD project! After having spent his masters project with us, he now comes back sponsored by the SOFI2 CDT. Sam will develop methods leveraging on convolutional neural networks to study protein conformational spaces. While currently Sam can only join us in an unusual “virtual” way, he is already at work. Looking forward to welcoming him in Durham, hopefully soon!

Scientist Next Door

In this period of lock down caused by COVID-19, children are home schooled. We have joined the “Scientist Next Door ” project, organised by scientists from the university of Edinburgh and Durham, to share our excitement for science with them!

We hope to hold group video calls with families and discuss things we find interesting, share ideas and resources. And, as Scientists Next Door,  after the lockdown is over, it would be great to meet in person!

If you are a scientist, consider joining too. If you are parent, get in touch! www.scientist-next-door.org

scientistnextdoor