Garvan Institute of Medical Research
Phenomics team within the Kinghorn Centre for Clinical Genomics
Post-doctoral research fellow position
The Garvan Institute of Medical Research is one of Australia’s leading medical research institutes, with over 600 scientists, students and support staff. We pioneer study into the most widespread diseases affecting our community today, including cancer, neurodegenerative and mental diseases, disorders of the immune system, diabetes and obesity, osteoporosis and other skeletal disorders.
The Kinghorn Centre for Clinical Genomics (KCCG) was established by the Garvan Institute as an Australian research and sequencing centre to deliver and interpret genome sequences for clinical use. KCCG is one of three initial world-wide sites to be acquiring an Illumina HiSeq X Ten sequencing factory, capable of sequencing 150 whole human genomes every 3 days. The KCCG facilitates genome-based research, particularly in cancer and monogenic diseases, but also in complex disease such as diabetes, osteoporosis and immunological disease. Our vision is to translate medical research into clinical care in Australia and beyond by integrating sequencing, bioinformatics and data management in a cutting-edge Genomics research environment.
The Phenomics Team within KCCG is a leading digital phenotyping group, with areas of expertise in modeling and acquiring phenotypes from unstructured data, cross-species integration of phenotype data and ontology-driven decision support for rare and complex disorders.
This position focuses on the research and development of innovative NLP / Machine Learning techniques to address the challenges of extracting multi-dimensional and longitudinal phenotypes from unstructured data.
The candidate will work alongside our clinical and scientific staff and will have the critical role to drive the phenotype-acquisition aspects of the centre’s patient data management platform. Some specific tasks will include but not be limited to: designing novel and innovative methods to extract canonical and non-canonical phenotypes and to associated them with orthogonal dimensions, such as degrees of severity or negation; analyze statistical patterns in large scale textual corpora to derive novel insights into the structure of phenotypes; publish high quality manuscripts in peer reviewed journals.
The candidate will have the opportunity to work with the team’s close collaborators from the Monarch Initiative, including groups from Charite Medical University Berlin, Oregon Health & Science University and Lawrence Berkeley National Laboratory.
Initially, the position will be awarded for 2 years.
- PhD in text mining, data mining, natural language processing, computational linguistics, computer science or related field
- Excellent knowledge in developing and adapting algorithms for text mining and machine learning
- Excellent Java skills
- Experience in biomedical/clinical Natural Language Processing/Text Mining
- Strong track record of high-quality papers in conferences such as ACL, EMNLP, Coling, ICDE, etc., and/or in high quality journals;
- High potential to develop independent research proposals
- Strong interpersonal and communication skills
- Flexible, proactive, creative and detail-oriented
Please send a statement of interest, an academic CV (in pdf format) and a first author publication to Tudor Groza (t.groza at garvan.org.au) with the subject line BIOPOSTDOC. Please do not include references - we may request these at a later stage. For informal queries, please send an email to t.groza at garvan.org.au.