@@ -58,6 +58,34 @@ Below we have collected some example responses in the formats required by [Wellc
<br><br>
</details> -->
### Wellcome Data Management Plan - Career Development Award (Dec 2022)
This is the data management plan for a fellowship application. It discusses sharing of a new sequence (called DIPPI), sharing of the developed tools, and sharing of the acquired data.
Prepared by Michiel Cottaar with feedback from Cassandra Gould van Praag and <mark>others</mark>.
<details>
<summary><b>Expand for application sections</b></summary><br>
I will make any research outputs for access and reuse following FAIR principles and the open science guidelines and tools from WIN (https://www.win.ox.ac.uk/open-win). This is in line with my previous software releases such as file-tree (https://pypi.org/project/file-tree/), pipe-tree (https://open.win.ox.ac.uk/pages/ndcn0236/pipe-tree/), and MCMRSimulator.jl (https://open.win.ox.ac.uk/pages/ndcn0236/mcmrsimulator.jl/dev/).
<br><br>
<details>
<summary><b>DIPPI Sequence</b></summary><br>
Sequence development requires access to Siemens' proprietary sequence development environment, which is governed by Oxford's Master Research Agreement with Siemens Healthineers. According to this agreement, any sequence transfer requires contracts between those sites, the University of Oxford, and Siemens. The WIN physics group has extensive experience negotiating such contracts in a timely manner and in sharing their MRI sequence innovations around the world. I will exploit this expertise (collaborators Prof. Miller and Dr. Wu) to share the DIPPI sequence with all parties who can meet the contractual requirements of the Master Research Agreement.<br><br>
MRI protocols, which describe suggested acquisition schemes for DIPPI, will be made available under a CC-BY 4.0 license through the WIN MR Protocol database (https://open.win.ox.ac.uk/protocols).
<br><br>
</details>
<details>
<summary><b>Data</b></summary><br>
MRI data is uniquely identifiable at most stages of processing and should be treated as personal data under GDPR. It is therefore not appropriate to share this data freely. WIN is developing a data portal which can provide gated access for bonafide researchers to GDPR sensitive data, and unrestricted access to anonymous data. Data shared on the portal will be deidentified as far as possible (for example by removing facial features and identifying metadata) while still retaining the research value of individual participant structure. Anonymous data and animal data will be shared on the portal under a CC-BY-4.0 license.GDPR sensitive data will be share with under terms of appropriate data usage agreement which will protect the privacy of participants and acknowledgement of the contribution. All participants will be asked for consent to share data in the way proposed, as described in the relevant ethical approvals.
<br><br>
</details>
<details>
<summary><b>Software</b></summary><br>
Software tools will be developed to fit the DIPPI model to estimate myelin thickness from DIPPI data. These tools will be included in a pipeline that can be used to fully process from raw DIPPI data to myelin maps. Once these tools are robust, they will be released as a python package on the WIN's own Gitlab page (https://git.fmrib.ox.ac.uk/) under a permissive open source license (MIT or Apache 2.0) and registered with both pypi and anaconda for easy installation. To improve discoverability, these tools would also be listed as affiliated with the FMRIB's Software Library (FSL; https://fsl.fmrib.ox.ac.uk/fsl/fslwiki/OtherSoftware), which is used by over a thousand labs around the world. This software will be given its own DOI using Zenodo. <br><br>
The MR simulator (MCMRSimulator.jl) will be developed in public following modern software engineering practices including regular releases, a change log, automated testing, and documentation for both users and developers (https://open.win.ox.ac.uk/pages/ndcn0236/mcmrsimulator.jl/dev/). Version 0.3 is already released under the Apache 2 permissive license. This will facilitate early discovery of any bugs not caught by the automated tests as well as maximising the chance of it being used by someone to discover new MRI probes of microstructure.
<br><br>
</details>
</details>
### Wellcome Data Management Plan - Discovery Platform (2022)
