Our Discovery Mission
Our mission, should we choose to accept it, is to discover and document all remaining Australian species of plants, animals, fungi and other organisms ... in a generation.
Part 3 - Whiteboards
The ideas below have been contributed to the whiteboard for discussion during the roundtable breakouts. You can add to them on the Whiteboard ideas page
Roundtable 6: What new technologies (other than genetics and genomics) will we need to deploy for our mission?
New technologies are available now that can help in our mission, including machine learning, high-end imaging, supercomputing and more. This roundtable will consider issues around new technologies, including:
How exactly could machine learning be used in species discovery, delimitation and documentation?
What support structures do we need to be able to effectively use new imaging technologies?
In reconstructing an Australian centric "Tree of Life" we will need to integrate new and novel ways to display very large phylogenetic trees with the ability to incorporate several different data layers. There are a few "Tree of Life" initiatives out there and tree viewing methods (opentreeoflife.org, onezoom.org) and there are taxa specific resources that could be mined (fishtreeoflife.org)
Improved automation of analysis of micro-CT data is needed. It seems like machine learning approaches could be used to 'train' a computer based on segmentation of a few specimens of a clade and then the software could handle the rest of the data.
There is a need for database information to be available with the specimens but exclude the laborious time spent adding labels, especially in insect collections. Pin head, for instance should be able to contain information through a QR system or barcodes that can be easily read.
Tabletop SEM, microCT scanning, pattern recognition software. Better national infrastructure funding for biological sciences to allow us to jump on or further develop these new technologies and add them to the workflow.
We require more collaboration and co-operation from information technology groups. The use of AI and pattern recognition software will save time for obtaining morphometrics of coral skeletons, and other taxa. Such software could even be used by ecologists to identify species if defined characters exist for species delineation.
Maximise re-use of descriptive data in all stages (single species description, monograph, "floras, faunas and fungas", identification products) through dispersed and standardised trait databases for scoring and mark up templates for end products. Avoid tedious re-formatting for specific end products/publications.
More (or better) ideas? Add them here.