In previous machine learning work it was noted that the low resolution limit to which diffraction data was integrated served as a strong feature when determining the chances of successful phase estimation. Initial electron density maps serve as a starting point for solvent flattening, MR or model building which ultimately result in the 3D structure of the protein of interest. The quality and reliability of these initial phases are therefore essential to boot-strap the final structure.
For this project we are looking for a candidate who can explore this topic using AI. Test diffraction data sets will be produced with varying low resolution limits and structure determination will be attempted by either MR or EP. The resulting electron density maps will then be evaluated regarding their information content, i.e. whether the details visible will allow for model building or not.
The diffraction data used is available from an in-house database, METRIX, which has been developed for machine learning purposes and the resulting electron density maps will have been created for this project in advance. The candidate will hence focus on training a cNN, ResNet or related alternatives in distinguishing interpretable maps from non-interpretable ones.
Diamond Light Source is the UK's national synchrotron science facility, located at the Harwell Science and Innovation Campus in Oxfordshire.
Copyright © 2022 Diamond Light Source
Diamond Light Source Ltd
Diamond House
Harwell Science & Innovation Campus
Didcot
Oxfordshire
OX11 0DE
Diamond Light Source® and the Diamond logo are registered trademarks of Diamond Light Source Ltd
Registered in England and Wales at Diamond House, Harwell Science and Innovation Campus, Didcot, Oxfordshire, OX11 0DE, United Kingdom. Company number: 4375679. VAT number: 287 461 957. Economic Operators Registration and Identification (EORI) number: GB287461957003.