Re-usability

Alrighty – so you have been learning about the Semantic Engine and how important documentation is when it comes to research data – ok, ok,  yes documentation is important to any and all data, but we’ll stay in our lanes here and keep our conversation to research data.  We’ve talked about Research Data Management and how the FAIR principles intertwine and how the Semantic Engine is one fabulous tool to enable our researchers to create FAIR research data.

But…  now that you’ve created your data schema, where can you save it and make it available for others to see and use?   There’s nothing wrong with storing it within your research group environment, but what if there are others around the world working on a related project?  Wouldn’t it be great to share your data schemas?  Maybe get a little extra reference credit along your academic path?

Let me walk you through what we have been doing with the data schemas created for the Ontario Dairy Research Centre data portal.   There are 30+ data schemas that reflect the many data sources/datasets that are collected dynamically at the Ontario Research Dairy Centre (ODRC), and we want to ensure that the information regarding our data collection and data sources is widely available to our users and beyond by depositing our data schemas into a data repository.   We want to encourage the use and reuse of our data schemas – can we say R in FAIR?

Storing the ADC data schemas

Agri-food Data Canada(ADC) supports, encourages, and enables the use of national platforms such as Borealis – Canadian Dataverse Repository.   The ADC team has been working with local researchers to deposit their research data into this repository for many years through our OAC Historical Data project.   As we work on developing FAIR data and ensuring our data resources are available in a national data repository, we began to investigate the use of Borealis as a repository for ADC data schemas.  We recognize the need to share data schemas and encourage all to do so – data repositories are not just for data – let’s publish our data schemas!

If you are interested in publishing your data schemas, please contact adc@uoguelph.ca for more information.   Our YouTube series: Agri-food Data Canada – Data Deposits into Borealis (Agri-environmental Data Repository) will be updated this semester to provide you guidance on recommended practices on publishing data schemas.

Where is the data schema?

So, I hope you understand now that we can deposit data schemas into a data repository – and here at ADC, we are using the Borealis research data repository.  But now the question becomes – how, in the world do I find the data schemas?  I’ll walk you through an example to help you find data schemas that we have created and deposited for the data collected at the ODRC.

1. Visit Borealis (the Canadian Dataverse Repository) or the data repository for research data.



2. In the search box type:  Milking data schema
3. You will get a LOT of results (152, 870+) so let’s try that one again
4. Go back to the Search box and using boolean searching techniques in the search box type: “data schema” AND milking



5. Now you should have around 35 results – essentially any entry that has the words data schema together and milking somewhere in the record
6. From this list select the entry that matches the data you are aiming to collect – let’s say the students were working with the cows in the milking parlour.  So you would select ODRC data schema: Milk parlour data

Now you have a data schema that you can use and share among your colleagues, classmates, labmates, researchers, etc…..

Remember to check out what you else you can do with these schemas by reading about all about Data Verification.

Summary

A quick summary:

1. I can deposit my data schemas into a repository – safe keeping, sharing, and getting academic credit all in one shot!
2. I can search for a data schema in a repository such as Borealis
3. I can use a data schema someone else has created for my own data entry and data verification!

Wow!  Research data life is getting FAIRer by the day!

Let’s take a little jaunt back to my FAIR posts.  Remember that first one?  R is for Reusable?  Now, it’s one thing to talk about data re-usability, but it’s an entirely different thing to put this into action.  Well, here at Agri-food Data Canada or ADC we like to put things into action, or think about it as “putting our money where our mouth is”.   Oh my!  I’m starting to sound like a billboard – but it’s TIME to show off what we’re doing!

Alrighty – data re-usability.  Last time I talked about this, I mentioned the reproducibility crisis and the “fear” of people other than the primary data collector using your data.  Let’s take this to the next level.  I WANT to use data that has been collected by other researchers, research labs, locales, etc… But now the challenge becomes – how do I find this data?  How can I determine whether I want to use it or whether it fits my research question without downloading the data and possibly running some pre-analysis, before deciding to use it or not?

ADC’s Re-usable Data Explorer App

How about our newest application?  the Re-usable Data Explorer App?  The premise behind this application is that research data will be stored in a data repository, we’ll use Borealis, the Canadian Dataverse Repository for our instance.  At the University of Guelph, I have been working with researchers in the Ontario Agricultural College for a few years now, to help them deposit data from papers that have already been published – check out the OAC Historical Data project.  There are currently almost 1,500 files that have been deposited representing almost 60 studies.   WOW!  Now I want to explore what data there is and whether it is applicable to my study.

Let’s visit the Re-usable Data Explorer App   and select Explore Borealis at the top of the page.  You have the option to select Study Network and Data Review.  Select Study Network and be WOWed.   You have the option to select a department within OAC or the Historical project.  I’m choosing the Historical project for the biggest impact!  I also love the Authors option.

Look at how all these authors are linked, just based on the research data they deposited into the OAC historical project!  Select an author to see how many papers they are involved with and see how their co-authors link to others and so on.

But ok – where’s the data?  Let’s go back and select a keyword.  Remember lots of files, means you need a little patience for the entire keyword network to load!!  Zoom in to select your keyword of choice – I’ll select “Nitrogen”.  Now you will notice that keywords needs some cleaning up and that will happen over the next few iterations of this project.  Alright nitrogen appears in 4 studies – let’s select Data Review at the top.  Now I need to select one of the 4 studies – I selected the Replication Data for: Long-term cover cropping suppresses foliar and fruit disease in processing tomatoes.

What do I see?
All the metadata – at the moment this comes directly from Borealis – watch for data schemas to pop up here in the future!  Let’s select Data Exploration – OOOPS the data is restricted for this study – no go.

Alrighty let’s select another study:  Replication Data for: G18-03 – 2018 Greens height fertility trial

Metadata – see it!  Let’s try Data exploration – aha!  Looking great – select a datafile – anything with a .tab ending – and you will see a listing of the raw data.  Check out Data Summary and Data Visualization tabs!

Wow!!  This gives me an idea of the relationship of the variables in this dataset and I can determine by browsing these different visualizations and summary statistics whether this dataset fits the needs of my current study – whether I can RE-USE this data!

Last question though – ok I’ve found a dataset I want to use – how do I access it?  Easy…  Go to the Study Overview tab scroll down to the DOI of the dataset.  Click it or copy it into your browser and it will take you to the dataset in the data repository and you can click Access Dataset to view your download options

Data Re-use at my fingertips

Now isn’t that just great!  This project came from a real use case scenario and I just LOVE what the team has created!  Try it out and let us know what you think or if you run into any glitches!

I’m looking forward to the finessing that will take place over the next year or so – but for now enjoy!!