|Funding for:||UK Students, EU Students|
|Funding amount:||Not Specified|
|Placed On:||3rd September 2019|
|Closes:||3rd December 2019|
Expected start date: 1st November 2019 (negotiable)
We have a fully funded PhD studentship available in the group to develop and apply text-mining and natural language processing (NLP) methods to biomedical literature and clinical records data to improve automated phenotype annotation and quantitative trait extraction. The project will focus on neuro-developmental diseases and will take forward our existing methods for integrating ontology based semantic analysis with clinical genetic data.
Overview of the Project
The aim of this project is to develop text-mining and natural language processing approaches to effectively extract relevant features from biomedical literature and clinical data records including behavioural, medical history, treatment, developmental milestones, growth, age and gender information for patients. This will involve using emerging techniques using vectors for word representation such as Word2Vec and GloVe and related approaches in an integrated ontology guided manner. We will investigate how linking molecular genotypic data with literature based phenotypic data can be used to quantitatively evaluate the relationships between these features at the level of patients. Such data structures promise to be powerful tools in stratifying patients by their features and defining more clearly the underpinning aetiology of disease. Informative features could form the basis for a wide range of downstream application areas; gene prioritisation, pathway analysis, mechanistic modelling, clinical profiling and diagnosis, evaluation of treatment options and efficacy and to identify new outcome measures for clinical trials. The project provides an excellent opportunity to apply computer science to a critically important and growing area of biomedical research.
The growing number of large cohort genetic studies for neuro-developmental diseases that include detailed phenotypic and clinical information represents a tremendous opportunity to bring to bear computer science and machine learning methods for joint analysis to derive new insights into disease aetiology and to inform future patient care. Several recent studies have illustrated the potential of such methods with particular interest in the use of text-mining with literature and electronic health records.
This is a fully funded 3.5-year position for UK or EU nationals.
We are seeking applications from highly motivated individuals with a background in computer science or computational biology especially those with expertise in NLP and an interest in applying their skills to literature and clinical records data. As a member of the Simpson research group and the Institute for Adaptive and Neural Computation at the School of Informatics you will benefit from a thriving research environment with strong PhD cohorts in Data Science and Biomedical AI. For more information please contact Ian Simpson directly.
Type / Role: