|
Overview
Programme
Details
Contact
|
|
  |
Overview
|
On Tuesday 20th June 2017, the Department of Mathematical Sciences at Durham University will hold a `Mini-Workshop' on
`Protein Structure Prediction and Modelling'
The programme for the meeting is shown below; the venue will be the Department of Mathematical Sciences, Durham University (map), in room CM 221.
Beside Protein Structure Prediction (PSP) and Protein Structure Modelling in a closer sense, also related topics such as the geometry of proteins, the evolution of proteins,
or the movement of proteins, fall into the remit of the workshop, ideally including a component of mathematical or statistical modelling.
|
|
Programme
| 14:00 | Georgios Karagiannis, Durham University
Parallel and Interacting Stochastic Approximation Annealing algorithms for global optimisation
We present the parallel and interacting stochastic approximation annealing algorithm,
a stochastic simulation procedure for global optimisation. The proposed algorithm is suitable
to address global optimisation problems in high dimensional or rugged scenarios, where standard optimization algorithms
suffer from the so-called local trapping problem. Central to our methodology is the idea of simulating a population of Markov chains
that interact each other in a manner able to overcome the local trapping problem. We demonstrate the good performance of the algorithm on a
theoretical protein folding application, and compare it with the performance of other competitors.
Karagiannis, G., Konomi, B.A., Lin, G., Liang, F. (2016), Parallel and Interacting Stochastic Approximation Annealing algorithms for global optimisation. Stat Comput.
|
| 14.30 | Chris Prior, Durham University
Folded curve interpretation of small angle scattering data
We use the continuous curve model of protein backbones to interpret small angle scattering data (Based on [1]).
Using expressions derived in [2] we "dress" this backbone with scattering centres for the amino acids and an explicit hydration layer in
order to obtain theoretical scattering curves; which are then fitted to scattering data. The idea behind the technique is that the
predictions resemble protein folds and can be easily assessed as viable by structural biologists. In addition the description has a
significantly reduced parameter space by comparison to existing small angle scattering ab-initio interpretative methods.
The technique "correctly" predicts the structure of known proteins (e.g. BSA,lysosome and ferritin derivatives), but there
will be a number of obstacles to general fitting which I would like to discuss.
[1] Hausrath, Andrew C., and Alain Goriely. "Repeat protein architectures predicted by a continuum representation of fold space." Protein Science 15.4 (2006): 753-760.
[2] Prior, Chris B., and Alain Goriely. "The Fourier transform of tubular densities." Journal of Physics A: Mathematical and Theoretical 45.22 (2012): 225208.
|
| 15:00 | Coffee
| |
| 15:30 | Daniel Bonetti, Instituto Federal São Paulo (IFSP)
Estimation of Distribution Algorithms for Protein Structure Prediction via k-means clustering
Proteins are essential for maintaining life. For example, knowing the structure of a protein, cell regulatory mechanisms of organisms can be modelled,
enabling disease treatments or relationships between protein structures and food attributes can be determined. However, we know that discovering the
structure of a protein is a difficult and expensive task that can cost five billion dollars and take 10 years just to figure out the cure of a specific
disease. Computational methods have been developed to find proteins structures. They require several calculations to predict even a small protein,
since it is hard to explore the large search. We developed an Estimation of Distribution Algorithm (EDA) specific for the ab initio Protein Structure
Prediction (PSP) problem using full-atom representation. We developed a multivariate probabilistic model to address the correlation among dihedral
angles of an EDA for PSP. We used the k-means clustering to find high density variable values in the search space. Then, we used these clusters
to generate the offspring of the evolutionary process. For each generation and correlate variables, a new k-means clustering algorithm is performed.
So, the k-means must create the clusters in a predefined amount of time. That ensures that the EDA does not spend too much time creating high quality models,
since an average model has the enough quality needed. Furthermore, we compared the proposed probabilistic model with k-means against Finite Gaussian Mixtures
and Multivariate Kernel Estimation.
|
| 16:00 | Jochen Einbeck, Durham University
The H2AX-histone as a radiation biomarker - modelling and dose estimation
Histones are a special type of proteins which `package' the DNA double helix to form chromatin. The H2AX histone has the property of getting phosphorylated following a DNA double-strand break (DSB). Since ionizing radiation leads to DSBs, H2AX histones can be used as a radiation biomarker. The phosphyrolation is manifested as fluorescent dots (γ-H2AX foci), which can be counted using immunofluorescence microscopy (or flow cytometers). Using in vitro laboratory data, with known source radiation, a statistical model can be built relating radiation dose to the number of foci. The estimated model can then be used, in principle, to produce a radiation dose estimate for a new given foci count, following (actual or suspected) exposure of an individual. A problem with this process is that the inter-individual variation is quite large, which, to some extent, renders in vitro calibration curves of limited use for in vivo dose estimation. Possible ideas to deal with this problem are outlined in this talk.
|
| 16:30 | Discussion and Close |
General Details
Registration and Submission
There is no registration fee.
For catering purposes it is appreciated if you let us know by June 13 if you would like to attend.
(But please free to show up spontaneously, or only for part of the workshop, as convenient.)
If you are interested in presenting your work at this workshop, please e-mail
us title and abstract.
Travel
The Department is easily accessible by bus, taxi or foot from the
railway station. More information about travelling to Durham can be found here. A useful map
can be found here.
Car parking at the Department of Mathematical Sciences is in extremely limited supply. For those arriving by car, it will be possible to obtain a day-permit at the entrance barriers - however this is no guarantee that you will find a place.
A preferable alternative would be to use the Durham Park and Ride at the nearby Howlands Park site which offers a large supply of secure parking.
Contact
To confirm attendance, or in case of any questions, please contact Jochen Einbeck, telephone (0191) 3343125.
Notes
This event is supported by Santander Universities UK.
|
