Nathan Harmston

Nathan Harmston

I am a BBSRC Case Phd student (co-sponsored by Unilever), co-supervised by Prof. Michael Stumpf and Wendy Filsell at Unilever. My undergraduate degree was in Computer Science from the University of Nottingham and I obtained a distinction in my Masters degree in Bioinformatics and Theoretical Systems Biology from Imperial College London.

I am investigating how text mining and natural language processing techniques can be applied to the biological literature to help with problems in Systems Biology. The literature contains a large amount of unstructured knowledge about biological systems.

Current research

  • distinguishing the species of named entities in biological/biomedical text
  • relation extraction using a variety of machine learning approaches
  • performing functional annotation of genes using the literature
  • automatically generating mathematical models of biological systems using the literature

I help organise the London Biogeeks meetings at Imperial College and organised the Elements of Statistical Learning book group.

Publications

Which species is it? Species-driven gene name disambiguation using random walks over a mixture of adjacency matrices.

Harmston N, Filsell W, Stumpf MP.

Bioinformatics. 2012 Jan 15;28(2):254-60. doi: 10.1093/bioinformatics/btr640. Epub 2011 Nov 30.

PMID:
22135416
[PubMed - indexed for MEDLINE]

Free Article

What the papers say: text mining for genomics and systems biology.

Harmston N, Filsell W, Stumpf MP.

Hum Genomics. 2010 Oct;5(1):17-29. Review.

PMID:
21106487
[PubMed - indexed for MEDLINE]

Free PMC Article

    Posters

    Disambiguating the Species of Named Entities using random walks over a mixture of adjacency matrices.

    Talks

    Integration of copy number, gene expression and pathways to investigate the ependymoma genome – UKAffy 2008

    Awards

    3rd place – Outstanding Second year PhD Research Poster – 2010