Developing A Roadmap For The Training Of Clinicians In Bioinformatics
News Oct 13, 2014
The multidisciplinary workshop was a key step in introducing clinicians to bioinformatics and addressing the question of how best to train them in the tools that are available within this field. It is essential to develop this training as, if clinicians understand bioinformatics; a plethora of genetic data becomes available to them. Not only will clinicians be able to assess and make use of genetic data that influences our lives and health, through an enhanced awareness of the processes involved, greater collaboration between clinicians and bioinformaticians can be achieved.
By bringing together experts in bioinformatics training and those with a clinical background, this workshop developed a roadmap to create and implement training. One outcome of the workshop was: it was agreed that training should be tailored to clinicians working in specific sectors due to limitations in time allocated to clinicians for training in new technologies. As well as a requirement for clinicians to understand the tools and semantics of bioinformatics, the group placed emphasis on the need to inform clinicians about the limitations of the data involved.
The workshop resulted in concrete, deliverable actions that set out the next steps for developing the training. These focused on gathering more information about the training needs of clinicians, the most valuable bioinformatics tools to achieve these needs and the most effective form for such training.
Vicky Schneider, Head of 361° Division (Scientific Training, Education & Learning) at TGAC, said: “What bioinformatics gives you is a lot of resources, databases, and tools to actually explore information related to clinical research which is openly available. For instance, someone working on the human genome may put their data in one repository but further information about parts of this genome may be deposited somewhere else. What you want is to be aware of where the all of the information is stored and the quality of this information. The initial findings arising from this workshop will be actioned at a follow-up workshop in Australia just weeks after this initial meeting. There isn’t one overall solution, but we do have a framework to think about the training and the training needs. It is exciting to start thinking about how we can do this training in UK and in Australia, and how we can benefit from sharing our experiences on what is working and what is not.”
Catherine Shang, Project Manager at Bioplatforms Australia, said: “The bottleneck to personalised medicine is no longer the sequencing technology; the limit now is a lack of expertise to analyse and interpret the large amounts of genomic data and to understand the clinical implications. This workshop is the first step toward developing a bioinformatics training program designed for the needs of the health and medical researchers. We aim to enhance research outcomes through improved bioinformatics skills, understanding of the public resources, databases available and data analysis tools to enable innovation in the ‘genomics age’.“
Ruth Lovering, Principal Research Associate at University College London (UCL), said: “From the discussions I have had with clinicians and research scientists at UCL, and at conferences, it is very apparent that there is a need to provide these experts with information not only about which databases are the most appropriate resources to use, but also how to use them and the limitations of the available datasets.”
The workshop is a collaboration between representatives from Anglia Ruskin University, Bioplatforms Australia, University College London, University of Cambridge and TGAC.
An artificial intelligence (AI) approach based on deep learning convolutional neural network (CNN) could identify nuanced mammographic imaging features specific for recalled but benign (false-positive) mammograms and distinguish such mammograms from those identified as malignant or negative.