News
See our Unsupervised Learning method "Denoising Autoencoder Self-Organising Map" (DASOM) although developed for the Early Cancer Diagnostics Project it is universally applicable to a wide range of applications. DASOM integrates autoencoders into a hierarchically organised hybrid model where the front-end component is a grid of topologically ordered neurons. The model maintains clustering properties but by extending and enhancing its visualisation capacity it enables an inclusion and analysis of the intermediate representative space. Download the paper for free at https://authors.elsevier.com/a/1X5v~3BBjKVz76
Prof Levitt one of the 2013 Nobel Prize laureates visited the Scientific Computing Research Unit in November 2017. Prof Levitt’s diverse interests have included RNA & DNA modelling, protein folding simulation, classification of protein folds & protein geometry, antibody modelling, x-ray refinement, antibody humanisation, side-chain geometry, torsional normal mode, molecular dynamics in solution, secondary structure prediction, aromatic hydrogen bonds, structure databases, and mass spectrometry.
We are thankful that Prof Levitt took time out of his busy schedule to engage with the students and trust he enjoyed discussing their research projects.
Public Lecture by Prof Michael Levitt, Stanford University
Birth & Future of Multiscale Modeling
Date: Friday 17th November, 2017
Venue: PD Hahn Lecture Theatre 3
Time: 15h30-16h30
Public Lecture available at http://media.uct.ac.za/engage/theodul/ui/core.html?id=4796c0df-188d-4220-bc7e-e221e5cf711a
Born in South Africa in 1947, Prof Levitt was awarded the 2013 Nobel Prize for Chemistry (shared with Martin Karplus and Arieh Warshel). Prof Levitt’s diverse interests have included RNA & DNA modelling, protein folding simulation, classification of protein folds & protein geometry, antibody modelling, x-ray refinement, antibody humanisation, side-chain geometry, torsional normal mode, molecular dynamics in solution, secondary structure prediction, aromatic hydrogen bonds, structure databases, and mass spectrometry. He has a passion in helping today’s young scientists gain the recognition and independence that his generation enjoyed.
In his talk, Prof Levitt will review how the development of multiscale models for complex chemical systems began in 1967. These same multiscale models have become increasingly popular in applications that range from simulation of atomic protein motion, to protein folding and explanation of enzyme catalysis.
October is Breast Cancer Awareness Month, which is an annual campaign to increase awareness of the disease
This is relevant to our research unit as we aim to classify cancer at the genetic and molecular (Glycan) level. Using our mechanistic understanding of glycosylation and glycolysis of tumours we design leads for molecular classes of Cancer. SCRU laboratories collaborate closely with medicinal laboratories as well as human biology laboratories to translate basic research findings into new clinical strategies for diagnosis and therapy.
Pages
- 1
- 2