Personal tools
You are here: Home Events ANC Workshop: Ian Simpson and Andreas Kapourani, Chair: Richard Shillcock

ANC Workshop: Ian Simpson and Andreas Kapourani, Chair: Richard Shillcock

— filed under:

What
  • ANC Workshop Talk
When Mar 14, 2017
from 11:00 AM to 11:00 AM
Where IF 4.31/4.33
Add event to calendar vCal
iCal

Ian Simpson

"Enabling multi-ontology annotation and analysis for disease research"

Very few biomedical data corpora are well structured and those that are, by design, often use bespoke annotation and analysis methods to derive insight from experimentally derived data. This makes it difficult to control the quality and maximise the coverage of annotation for the useful integration of domain knowledge. This problem is exacerbated by the explosion in the number of potentially useful biological ontologies, now in excess of 500, many of which have no or only sparse annotation. In this talk I will describe some of our work developing a standardised ontology annotation framework for biological data and a suite of downstream analytic tools to aid in their use for downstream analysis. I will finish by demonstrating the concept of a "disease environment" that aims to capture the integrated domain knowledge in a formal graph structure and highlight some active areas of future work.



Andreas Kapourani

''Probabilistic modelling of DNA methylation profiles''

DNA methylation is an intensely studied epigenetic mark, yet its functional role is incompletely understood. Attempts to quantitatively associate average DNA methylation to gene expression yield poor correlations outside of the well-understood methylation-switch at CpG islands.

Here we use probabilistic machine learning to extract higher order features which quantitate precisely notions of shape of a methylation profile, capturing spatial correlations in DNA methylation across genomic regions. Using these features across promoter-proximal regions, we are able to construct a powerful machine learning predictor of gene expression, significantly improving upon the predictive power of average DNA methylation levels. Our results support previous reports of a functional role of spatial correlations in methylation patterns, and provide a mean to quantitate such features for downstream analyses.