Centre for Computational Science
Biological and Biomedicine Research highlights
David Wright, Maiki Cwiok, Shunzhou Wan, Hugh Martin, Rupert Nash, Hywel Carver, Gary Doctors, & Peter Coveney.
Centre for Computational Science, Department of Chemistry, UCL, 20 Gordon Street, London WC1A 0AJ.
Research in the Centre for
Computational Science (CCS) spans a number of biological and biomedical areas,
including: HIV, Cancer, Gene translocation, and Cerebral blood flow.
The emergence of drug resistant strains of HIV represents a major challenge in the treatment of patients who contract the virus. Currently clinicians use statistically derived clinical decision support systems (CDSS) to assess the resistance profile of the viral strain infecting each patient in order to determine the optimal treatment regimen. We aim to extent this by using predictive molecular dynamics (MD) simulations of the main target enzymes: protease, reverse transcriptase and integrase. Free energy calculations allow us to provide both quantitative and qualitative insight into the molecular causes of resistance.
Lung carcinoma, like all other forms of cancer, has two main characteristics: uncontrolled growth of cells and the ability of these cells to spread to distant sites. The epidermal growth factor receptor (EGFR) is a major target for drugs in treating lung carcinoma. Mutations have been correlated with EGFR activity and drug sensitivity. We have performed two studies: using multiple (ensemble) short MD simulations to investigate the binding affinities of tyrosine kinase inhibitors to wild-type and mutant EGFRs, which shows that ensemble simulations correctly rank the binding affinities for the inhibitor-EGFR systems; and the use of long timescale MD simulations to study mutation’s effect on EGFR activation, which reveals that the EGFR mutation promotes conformational changes, changing the relative preference between the active and inactive conformations and hence the activation of the EGFR kinase.
The translocation of polynucleotides through transmembrane protein pores is a fundamental biological process with important technological and medical relevance. Our research explores the use of two MD simulation translocation methodologies for the nucleotide-nanopore translocation system, with α-hemolysin as the nanopore: a) the use of non-equilibrium constant velocity-steered MD simulations of nucleic acid molecule translocation through the protein nanopore α-hemolysin and the use of Jarzynski's identity to determine the associated free energy profiles; b) the application of an adaptive biasing force, which can also be used to calculate free energy profiles. This approach is used to explain the observed differences in experimental translocation time through the nanopore between polyadenosine and polydeoxycytidine. Our simulations provide insight into the role of the interactions between the nucleic acid molecules and the protein pore.
Cerebral Blood Flow/Physiology
Cerebral blood plays an important role in development of neurovascular pathologies such an intercranial aneurysms, although the details are not well understood. We use ﬂuid ﬂow simulation in combination with patient speciﬁc medical imaging data in order to not only help understand blood ﬂow patterns in the presence of neurovascular pathologies, but also to provide a tool which surgeons can use in diagnostic and therapeutic capacities. To be clinically relevant, such a tool must be able to rapidly return reliable results to surgical teams while they are planning interventions, stressing the need for high performance computing and novel strategies such as in situ rendering.