News Release

Applying machine learning to biomedical science

How deep learning and ensemble methods are working together

Peer-Reviewed Publication

University of Sydney

Dr Pengyi Yang

image: Dr Pengyi Yang from the Charles Perkins Centre and School of Mathematics & Statistics at the University of Sydney. view more 

Credit: University of Sydney

With potential application diagnosing cancer or predicting how viruses, such as HIV, attack human cells, machine learning is opening promising new areas of application for bioinformatics - the data science of molecular biology. Dr Pengyi Yang from the Charles Perkins Centre and School of Mathematics and Statistics with colleagues has summarised the latest developments in this emerging field in a review article in Nature Machine Intelligence.

Latest techniques are bringing together two previously disparate approaches to machine learning: ensemble methods and deep learning.

Just like 'many heads are better than one', ensemble deep learning combines multiple 'computer brains' to achieve high levels of performance. Dr Yang summarises the latest developments in ensemble deep learning and its application in a range of biological and biomedical fields; highlights achievements unattainable by traditional methods; and maps out its potential to revolutionise molecular biological and biomedical sciences.


'Ensemble deep learning in bioinformatics', Nature Machine Intelligence
DOI: 10.1038/s42256-020-0217-y
Authors: Yue Cao, Thomas Andrew Geddes, Jean Yee Hwa Yang, Pengyi Yang

Corresponding author:

DECLARATION: The authors receive funding from the Australian Research Council, National Health and Medical Research Council and the University of Sydney.

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.