Wrapping up the MyTardis installations at Swinburne

Last week we finalised Swinburne's MyTardis installations and dealt with a few remaining tasks. Hopefully we'll have production systems up and running before Christmas. Although we don't use the actual  blue Police Box the concept is, not surprisingly, similar. MyTardis was developed for microscopy researchers at  the Monash University eResearch Centre to solve the long-standing technical challenge of securely storing and sharing large research datasets using an online system.  By integrating seamlessly with individual scientific instruments and laboratory or university research storage, My Tardis enables researchers to store, archive, and access data efficiently and most importantly without hassle. 

As part of the Swinburne Metadata Stores Project,  customised MyTardis installations were developed for researchers at the Brain and Psychological Sciences Research Centre (BPsyC), and the Centre for Atom Optics and Ultrafast Spectroscopy (CAOUS) – the centre has since been renamed as the Centre for Quantum and Optical Science (CQOS). This project was a collaboration between our team in Swinburne Research and the Victorian eResearch Strategic Initiative (VeRSI).

For BPsyC researchers, a MyTardis pilot system was developed specifically for the Elekta TRIUX MEG instrument. Magnetoencephalography (MEG) is a safe, non-invasive human brain imaging technique. The MEG scanner measures the very small magnetic fields produced by the active brain using external sensors configured as a hemispherical grid that covers the head. Applications of MEG include basic research into perceptual and cognitive brain processes, localising regions affected by pathology before surgical removal, and determining the function of various parts of the brain. Very cool. The customised My Tardis system captures images/data from the instrument's staging area (an institutional research data storage facility) via an Atom feed, creates tabled metadata (technical/scientific metadata) using customised metadata filters, and produces thumbnails and diagnostic plots that researchers can view to quickly identify useful data.

For CAOUS, a pre-configured MyTardis system developed specifically for microscopy data provides and storage of data and tabled metadata from the Olympus IX81 Confocal Microscope.Basic data capture and storage was also set up for the Renishaw In Via Raman Spectrometer and Bruker Contour GT-K1 Optical Surface Profiler. 

In future (and with a little more TLC!)  we hope that the systems will mature, and become more widely used by researchers as part of their day-to-day activities. So far Monash University, RMIT, the University of Sydney, and the University of Queensland have  implemented versions of MyTardis as an institutional data management solution. The Australian Synchrotron deployment has captured over 20 terabytes and 10,000 datasets!, archiving and providing data access and sharing tools to its community of thousands. Monash have used the system to make astronomy datasets publicly available. Although the metadata description implementation is a little clunky, it's a great example of what My Tardis is capable of and demonstrates the value of a data archiving system that provides separate instrument parameters and thumbnail images. Astronomy imaging data naturally fits into this type of system by virtue of its FITS header format.  Below is a screenshot of Atlas of Galaxy Spectral Energy Distributions From The UV to the Mid-Infrared (Principal Investigator: A/Prof. Michael Brown, Monash University)

  Atlas of Galaxy Spectral Energy Distributions From The UV to the Mid-Infrared  (Principal Investigator: A/Prof. Michael Brown, Monash University)

Atlas of Galaxy Spectral Energy Distributions From The UV to the Mid-Infrared (Principal Investigator: A/Prof. Michael Brown, Monash University)

the wonderful [new] world of .astronomy

I first heard of .Astronomy last year when I was at Oxford university. My friend Phil Marshall had just returned from (or was about to head off to?) the fourth conference, aptly named .Astronomy 4. At the time I knew that it involved hackathons between Galaxy Zoo, SpaceWarps, Zooniverse folks, and perhaps others. But I did not know much else. It’s possible that early on it didn’t have the same web presence it clearly has now, or perhaps a reflection of my own ignorance. I did know that Phil and quite a few other Oxford Astros had quite a few more programming languages up their sleeves, a much better knowledge of APIs, Analytics, and data visualization tips and tricks.

  Simpson et al. (2012)

Simpson et al. (2012)

Following that conference, Bruce Berriman from IPAC – the Infrared Processing and Analysis Center at California Institute of Technology – wrote a really nice summary of highlights on his Astronomy Computing blog. The full "unproceedings" (screenshot bellow can be downloaded from astro-ph. As far as I'm aware this is the only .Astronomy proceedings. No formal proceedings exists for earlier or later conferences (so far..)

Last year’s conference had ~ 50 participants, astronomers from various research groups around the world, and a smattering software developers from companies including Microsoft. This years conference (.Astronomy 5)  was held in September, at the Microsoft New England Research & Development Center (NERD). It included a mix of hack day sessions and talks about eResearch, data visualisation, and communication strategies. Topics that I’ve since become more aware of and come across in my job. You can still check out their Live blog.

If thats not impressive enough, in July this year they launched a new journal - Astronomy & Computing - "intended to serve the community sitting, sometimes slightly awkwardly, between those two fields". What I love most about .Astronomy is that it actively encourages blue-sky thinking, playtime (inevitable creation of sandboxes) and innovation. Hopefully I'll have the chance to participate in .Astronomy 6!

Oh and they keep up a really great blog...

The MyTardis and MicroTardis Project

One component of ANDS Metadata Stores Project involves the capture of brain imaging metadata, through the implementation of Tardis software linked to the Brain and Psychological Sciences Research Centre's (BPsyC) state of the art instrumentation. Customisation of the new BPsyC-Tardis system is a collaboration between Swinburne University and VeRSI, and addresses the specific needs of BpsyC researchers. The sensitive nature of the data and the legislation governing data retention often restricts data sharing in this field. Consequently the main driver for BPsyC-Tardis is the need for secure data storage, while still allowing collaborative sharing. Phase I development facilitates automatic capture, initial processing and storage of magnetoencephalographgy (MEG) non-invasive 'brain scanning' data, and the accompanying external metadata - for example audio and visual trigger information.

A similar system was implemented for the Centre for Quantum Optics and Science (CQOS - formerly CAOUS). Phase I development of CAOUS-Tardis facilitates the collection, initial processing and storage of data from the Renishaw In Via Raman Spectrometer and Olympus IX81 Confocal Microscope, and the their accompanying external metadata; for example information about cell and bacteria samples. Phase II development will include data ingestion from the Bruker Contour GT-K1 Optical Surface Profiles.

Ok, so it's not in anyway related to THE TARDIS (insert gratuitous picture on the right), but the concept is similar. MyTardis makes it simple to store and share seemingly large datasets with collaborators inside or external to their institution, with an easy to use sharing interface. Later on, they can open up access to their data, providing a DOI in their research paper for anyone to access. This approach has seen data in MyTardis referenced in journals such as Science, Nature, PNAS, Plos One and more.

MyTardis was developed by researchers and software developers at Monash University: Androulakis S, Schmidberger J, Bate MA, Degori R, Beitz A, Keong C, Cameron B, McGowan S, Porter CJ, Harrison A, Hunter J, Martin JL, Kobe B, Dobson RC, Parker MW, Whisstock JC, Gray J, Treloar A, Groenewegen D, Dickson N, Buckle AM. (2008) Federated repositories of X-ray diffraction images. Acta Crystallogr D Biol Crystallogr. Jul;64(Pt 7):810-4.

Woodland Trust and A Year in the Wild

Last week the Guardian posted the finalists from this year's Woodland Trust photography competition. The photos really are beautiful.  The trust is UK leading woodland conservation charity, caring for over 1000 woods, roughly 50,000 acres (20,000) hectares of land. For Australia this sounds tiny, especially since our cattle ranches are typically larger than that. But the biodiversity of these woodlands is phenomenal.

If you love nature documentaries, make sure you check out BBCs A Year in the Wild, a documentary series exploring a year in Britain's most iconic national Parks: Swowdonia, the New Forest, and Cairngorms. Personally I think Snowdownia is one of the most beautiful places on this earth. Time to buy a new DSLR and start photographing the critters in the garden.


The Swinburne/ANDS Metadata Stores Project

In late 2012, Swinburne was successful in obtaining funding from the Australian National Data Service (ANDS) for the development of MyTardis and ReDBox research data management systems. The overall goal of the initiative was to implement a number of systems with the potential to support university-wide solutions for the discovery, sharing and re-use of rich data collections. Swinburne's research data collections would also be fed directly to Research Data Australia, the primary data discovery service of the Australian National Data Service. There area many potential benefits of such systems including improved research data management and compliance, increased visibility of Swinburne's research projects, data-discovery and the ability to assess needs and contribute to strategic planning decisions concerning ITS infrastructure for research.