Seminar with Mikael Boden


Title: Protein traffic control

Speaker: Mikael Boden
Time: 12-May, 11:00am
Place: Room 78-621/622

The cell is a decentralised but still carefully controlled device, shuttling
gene products, like proteins, through various compartments where they
perform their functions. Mechanisms for this protein traffic control are not
yet fully understood and machine learning techniques are being utilised to
discover protein sequence signals that allow targeted drug design and the
development of models that can be used to automatically annotate the growing
number of sequences that are yet to be experimentally characterised.
In relation to the current state-of-the-art, we report on improvements in
accuracy for a range of models for predicting localisation of proteins --
making use of advanced machine learning techniques like recurrent neural
networks and support vector machines. In particular we focus on the
essential compartments and organelles, the secretory pathway, the
mitochondrion, the chloroplast and the peroxisome. Our models are housed in
the Protein Prowler, our online sequence prediction server, which has been
endowed with graphical features allowing the user to scrutinize the outputs
of the models.









World-class basic and applied inter-disciplinary research on questions fundamental to understanding, designing and managing complex systems
2009 The ARC Centre for Complex Systems, Australia