FAIR-publishing.qmd

---
title: "How to publish FAIR research outputs"
author: "Eirini Zormpa"
institute: "The Alan Turing Institute"
title-slide-attributes:
    data-background-image: images/RSC_STACKED-LOGO-WHITE.png, images/SSI_SUBMARK-LOGO-WHITE.png,  images/rsf-logo-white.png
    data-background-size: 50px  50px
    data-background-position: top 20px left 50px, top 20px left 120px, top 20px left 190px
    data-background-opacity: "0.8"
format:
  revealjs:
    theme: [moon, custom.scss]
    logo: images/rsf-logo-white.png
    slide-numner: true
from: markdown+emoji
---


## Hi, I'm Eirini :wave:

![](https://www.turing.ac.uk/sites/default/files/styles/people/public/2022-06/eirini-zormpa.jpg?itok=Cx2rNcHY){.absolute top=100 right=50 width="320" height="384"}

::: {.column width="60%"}
**Eirini Zormpa**, Community Manager Open Collaboration @ AIM RSF
:::

## AI for Multiple Long-term Conditions Research Support Facility (AIM RSF)

![](images/AIM-RSF_explanatory-image.jpg){fig-align="center"}


## Hi, I'm Eirini :wave:

![](https://www.turing.ac.uk/sites/default/files/styles/people/public/2022-06/eirini-zormpa.jpg?itok=Cx2rNcHY){.absolute top=100 right=50 width="320" height="384"}

::: {.column width="60%"}
**Eirini Zormpa**, Community Manager Open Collaboration @ AIM RSF

<br>

Previously:

- Trainer on Research Data Management & Open Science @ Delft University of Technology :computer:
- Doctoral Researcher @ Max Planck Institute for Psycholinguistics :brain:
:::


# Why and what to share{background-color="#F26B5E"}

## Advertising scholarship

> An article about computational science in a scientific publication is not the scholarship itself, it is merely **advertising** of the scholarship.

::: {.fragment .fade-down}
Jon Claerbout, paraphrased in Buckheit & Donoho (1995) ([link](https://doi.org/10.1007/978-1-4612-2544-7_5), [free version](https://statweb.stanford.edu/~wavelab/Wavelab_850/wavelab.pdf)).
:::

::: aside
Slide credit: [Dr Anna Krystalli](https://annakrystalli.me/talks/reproducibility-value-in-practice.html#13)
:::


## The actual scholarship

The actual outputs created during a research project will differ but some examples include:

- Preregistrations
- Presentations and posters on your findings
- Datasets
- Protocols and workflows
- Code (analysis scripts and software packages)
- Other outputs, e.g. educational materials

## The actual scholarship

The actual outputs created during a research project will differ but some examples include:

- Preregistrations
- Presentations and posters on your findings
- **[Datasets]{style="color:#F2B705"}**
- Protocols and workflows
- **[Code (analysis scripts and software packages)]{style="color:#F2B705"}**
- Other outputs, e.g. educational materials

. . .

**You don't know what will be useful for others!**

## Example

![](https://www.turing.ac.uk/sites/default/files/styles/people/public/2018-06/evelina-gabasova_1000x921_jpg.jpg?itok=SUNC-Irv){.absolute top=100 right=50 width="320" height="384"}

::: {.column width="60%"}

**Evelina Gabasova**, Principal Research Data Scientist

::: {.r-stack}
![](images/evelina-github.png){.fragment}

![](images/zenodo-sw.png){.fragment}

![](images/gs-sw.png){.fragment}
:::

:::

## Why share all these other outputs?

![](images/share-work.jpg)


## What (not) to share

- Personal data is tricky - some data can be shared with proper consent and (pseudo)anonymisation.

. . . 

- Confidential data that's commercially sensitive or relevant for national security - talk to the data provider about what you **can** share.

. . . 

- Really big data (e.g. in quantum computing, cosmology) - think about what data is **necessary** and whether the data can be recreated.

. . .

- Anything that you found online that isn't licensed for reuse - *more on licences later*!


# How to share{background-color="#F26B5E"}

## The FAIR principles

![](images/FAIRPrinciples.jpg){fig-align="center"}


## Findable: Metadata :mag:

First people need to be able to find your outputs!

. . .

For outputs to be findable, they need to be described with **rich metadata**.
These metadata can be generic (e.g. title, author name, keywords) or discipline-specific.

<br>

. . .

### Resources
- [FAIRsharing.org](https://fairsharing.org/): a curated resource to help you find (among others) metadata standards

. . .

- [CodeMeta generator](https://codemeta.github.io/codemeta-generator/): a tool to create a minimal metadata file for software



## Findable: Persistent identifiers :mag:

Outputs should also be assigned a **unique and persistent identifier**, e.g. a Digital Object Identifier (DOI).
This makes it easy to find outputs, but also to link them with other relevant information (e.g. a publication).

![](images/DOI.jpg){fig-align="center"}


## Findable: Persistent identifiers for people :mag:

Persistent identifiers for researchers help if you have a common name or if you change your name!

![](images/OrcidIDs.jpg){fig-align="center"}


## Task: Create an ORCID iD {background-color="#F2B705"}

```{r}
library(countdown)
countdown(minutes = 3)
```

:alarm_clock: **3 minutes**

1. Navigate to [https://orcid.org/](https://orcid.org/) on your browser
2. Select Sign in/Register and then Register as a new user
3. Fill in the information required


## Accessible: Define processes :door:

After people have found your outputs they need to be able to access them!

- :unlock: This could mean that they are openly and freely available in a **repository**.

. . .

- :lock: Sensitive information may not allow for research outputs to be freely shared; in those cases the **access method should be clearly and transparently described** and the metadata should still be shared.

. . . 

- :x: If the output is no longer usable, the **metadata should still be accessible**.


## Interoperable :gear:

It's important to make all your outputs available in **open file formats**, that anyone can open and edit.

. . .

<br>

Using **controlled vocabularies** is also highly recommended, if these exist in your field.

## In how many different ways can you say "female"? {.smaller}

:::: {.columns}

::: {.column width="25%"}
18-day pregnant females <br>
female (lactating) <br>
individual female <br>
worker caste (female) <br>
2 yr old female <br>
female (pregnant) <br>
sex: female <br>
400 yr. old female <br>
female (outbred) <br>
mare <br>
female (other) <br>
adult female <br>
female parent <br>
female (worker) <br>
female plant
:::

::: {.column width="25%"}
femal <br>
castrate female <br>
female with eggs <br>
ovigerous female <br>
3 female <br>
cf.female <br>
female worker <br>
oviparous sexual females <br>
female (phenotype) <br>
cystocarpic female <br>
gynoecious <br>
thelytoky <br>
dikaryon <br>
female virgin <br>
dioecious female

:::

::: {.column width="25%"}
femlale <br>
female (gynoecious) <br>
remale <br>
metafemale <br>
f <br>
femele <br>
sterile female <br>
famale <br>
normal female <br>
femail <br>
sf <br>
female <br>
females <br>
tetraploid female <br>
strictly female
females only <br>
:::

::: {.column width="25%"}

worker <br>
hexaploid female <br>
healthy female <br>
female (gynoecious) <br>
probably female <br>
female (note: this sample was originally provided as a "male" sample and labeled this way in the paper and original submission; however analyses carried out in the meantime clearly show that this sample stems from a female individual)

:::


::::

::: aside
Adapted from Silvester *et al.*, 2015.
:::

## In how many different ways can you say "female"? {.smaller}

:::: {.columns}

::: {.column width="25%"}
18-day pregnant females <br>
female (lactating) <br>
individual female <br>
worker caste (female) <br>
2 yr old female <br>
female (pregnant) <br>
sex: female <br>
400 yr. old female <br>
female (outbred) <br>
mare <br>
female (other) <br>
adult female <br>
female parent <br>
female (worker) <br>
female plant
:::

::: {.column width="25%"}
[femal]{style="color:#F2B705"} <br>
castrate female <br>
female with eggs <br>
ovigerous female <br>
3 female <br>
cf.female <br>
female worker <br>
oviparous sexual females <br>
female (phenotype) <br>
cystocarpic female <br>
gynoecious <br>
thelytoky <br>
dikaryon <br>
female virgin <br>
dioecious female

:::

::: {.column width="25%"}
[femlale]{style="color:#F2B705"} <br>
female (gynoecious) <br>
[remale]{style="color:#F2B705"} <br>
metafemale <br>
f <br>
[femele]{style="color:#F2B705"} <br>
sterile female <br>
[famale]{style="color:#F2B705"} <br>
normal female <br>
[femail]{style="color:#F2B705"} <br>
sf <br>
female <br>
females <br>
tetraploid female <br>
strictly female
females only <br>
:::

::: {.column width="25%"}

worker <br>
hexaploid female <br>
healthy female <br>
female (gynoecious) <br>
probably female <br>
[female (note: this sample was originally provided as a "male" sample and labeled this way in the paper and original submission; however analyses carried out in the meantime clearly show that this sample stems from a female individual)]{style="color:#F2B705"}

:::


::::

::: aside
Adapted from Silvester *et al.*, 2015.
:::

## Dates

To avoid ambiguity, use the [RFC3339](https://datatracker.ietf.org/doc/html/rfc3339) standard: **YYYYMMDD**.

![](https://upload.wikimedia.org/wikipedia/commons/thumb/3/3a/Date_format_by_country_revised.svg/1600px-Date_format_by_country_revised.svg.png){width=900 height=500}

::: aside

This [image](https://en.m.wikipedia.org/wiki/File:Date_format_by_country_revised.svg) was created for [Wikimedia Commons](https://commons.wikimedia.org/wiki/Main_Page) and is used under a [CC-BY-SA licence](https://creativecommons.org/licenses/by-sa/4.0/).

:::


## Reusable: Documentation :recycle:

To be able to reuse your work, people need to be able to *understand* it.
This means you need to provide good **documentation**:

. . .

<br>

At a minimum, you should provide a **README** file where you describe what the project is about, how the files are organised and how to reproduce the project

## Example of a [README file]((https://www.zenodo.net/record/1411479#.ZEY_V3bMIV-))

![](images/zenodo-sw.png)


## Reusable: Usage licences :recycle:

You can't just use everything that's online; the creator* of the work holds the copyright to it!

::: aside
*In fact, it is often your university or institute that holds the copyright to the work you created while working there.
:::

. . .

You need to **tell people what they can do** with your work by providing a licence.

. . .

Usage licences are different for data and for code:

- Commonly used licences for data, presentations, papers etc. are the [Creative Commons licences](https://creativecommons.org/about/cclicenses/)
- For code, use [OSI-approved licences](https://choosealicense.com/licenses/)

## Licence example

![](images/zenodo-ttw.png)

## Creative Commons (CC) licences

![](https://upload.wikimedia.org/wikipedia/commons/a/a5/CC_License_Freedom_Scale_Chart.png?20161006115208){fig-align="center"}

::: {style="font-size: 60%;"}
Source: [How to attribute Creative Commons photos by Foter](https://foter.com/blog/how-to-attribute-creative-commons-photos/)
:::

## OSI-approved licences for code

::: {.r-stack}
![](images/tldrlegal1.png){.fragment}

![](images/tldrlegal2.png){.fragment}

:::

::: aside
[tl;drLegal](https://www.tldrlegal.com/license-tags/osi-approved)
:::

## How to apply licences?

:warning: **Check whether your university has a policy on sharing and licensing your research outputs.** :warning:

<br>

. . .

1. Provide the full text of the licence in your repository.

. . .

2. State under what licence your work is shared in the README file.

. . .

3. Ideally, include a [CITATION.cff file](https://citation-file-format.github.io/) in your repository and state in your README how you would like to be cited.

# Where to share?{background-color="#F26B5E"}

## Where to share: Supplementary materials

:::: {.columns}

::: {.column width="50%"}

#### Pros :thumbsup:

- Convenient for author
- One output to cite (?)

:::

::: {.column width="50%"}

#### Cons :thumbsdown:

- Can be hard to find
- Not suitable for all outputs (in terms of privacy, file format, relevance)
- Not very FAIR (no separate PID, licence)
- Sustainability is unclear

:::

::::

## Where to share: Repositories

Online storage spaces for sharing research data and other outputs: they can be generic or specific to a discipline or institute.

:::: {.columns}

::: {.column width="50%"}

#### Pros :thumbsup:

- Provide a DOI or other PID
- Require metadata
- Require a usage licence
- Encourage documentation
- Encourage open file formats
- Continued access

:::

::: {.column width="50%"}

#### Cons :thumbsdown:

- Require more knowledge (e.g. to pick a repository, licence etc.)
- Could initially cause confusion in terms of whether the publication or this deposisition should be cited.

:::

::::


## Discipline-specific repositories

Where possible, it's a good idea to use a disclipline-specific repository, usually for data and publications.

. . . 

- The metadata standard is likely to be tailored to the data you work with

. . .

- It is possible that there are restricted access options to account for privacy concerns.

. . .

- Fewer people may see your data, but they're more likely to be interested in it.

## How to find repositories?

![](images/re3data.png)

::: aside
[Registry of research data repositories](https://www.re3data.org/)
:::


## Data

::: {.r-stack}

![](images/gwas.png){.fragment}

![](images/earthchem.png){.fragment}

![](images/idr.png){.fragment}

![](images/uk-data-service.png){.fragment}

:::


::: aside
[GWAS Catalog](https://www.ebi.ac.uk/gwas/), [EarthChem](https://www.earthchem.org/), [Image Data Resource](http://idr.openmicroscopy.org/about/), [UK Data Service](https://ukdataservice.ac.uk/)
:::

## Protocols

![](images/protocols.png)

::: aside
[protocols.io](https://www.protocols.io/)
:::


## Generic repositories

A domain-specific repository is not always available, or it may not make sense for your outputs.
In that case there are generic repositories that are domain-agnostic and which accept a broad range of outputs (e.g. from data to illustrations).

## Zenodo

[Zenodo](https://zenodo.org/) is an open repository that accepts most research outputs.

- :white_check_mark: Funded by CERN, OpenAIRE, and the European Commission.
- :white_check_mark: Built on open source infrastructure.
- :white_check_mark: Offers integration with GitHub to archive code.

![](images/zenodo.png)

## Demo of syncing a GitHub and Zenodo repository {background-color="#F2B705"}

Just watch me for now :computer::eyes:

## Picking a repository

:::: {.columns}

::: {.column width="40%"}
![](images/trust-principles.png)
:::

::: aside
Source: [The TRUST Principles - An RDA Community Effort](https://www.rd-alliance.org/trust-principles-rda-community-effort)
:::

::: {.column width="60%"}
![](images/trust-principles-explained.png)

:::

::::


## Where to share: Data or software paper

For data or software of special importance, you may consider writing a data or software article.


:::: {.columns}

::: {.column width="50%"}

#### Pros :thumbsup:

- Additional publication
- More space to dig into methodology than in a traditional article

:::


::: {.column width="50%"}

#### Cons :thumbsdown:

- More time-consuming

:::

::::


## Example of a data paper

![](images/scientific-data-example.png)

::: aside
[scientific data](https://doi.org/10.1038/s41597-023-02125-y)
:::

## Example of a software paper

![](images/joss-example.png)

::: aside
[Journal of Open Source Software](https://doi.org/10.21105/joss.05153)
:::

## Summary

- There are more research outputs worth publishing than the traditional paper!
- These outputs should be published in accordance with the FAIR principles.
- Outputs don't need to be open to be FAIR.
- Repositories (discipline-specific or generic) are a great place to publish your outputs.
- Data or software papers may be suitable for more impactful outputs.

## Thank you!

![](images/thank-you.jpg){fig-align="center"}

## References {.smaller}
- AIM RSF, & Scriberia. (2023). Illustrations for AI for multiple long-term conditions: Research Support Facility. Zenodo. [https://doi.org/10.5281/zenodo.8042780](https://doi.org/10.5281/zenodo.8042780)
- Buckheit, J.B., Donoho, D.L. (1995). WaveLab and Reproducible Research. In: Antoniadis, A., Oppenheim, G. (eds) Wavelets and Statistics. Lecture Notes in Statistics, vol 103. Springer, New York, NY. [https://doi.org/10.1007/978-1-4612-2544-7_5](https://doi.org/10.1007/978-1-4612-2544-7_5)
- Chue Hong, Neil P., Katz, Daniel S., Barker, Michelle, Lamprecht, Anna-Lena, Martinez, Carlos, Psomopoulos, Fotis E., Harrow, Jen, Castro, Leyla Jael, Gruenpeter, Morane, Martinez, Paula Andrea, Honeyman, Tom, Struck, Alessandra, Lee, Allen, Loewe, Axel, van Werkhoven, Ben, Jones, Catherine, Garijo, Daniel, Plomp, Esther, Genova, Francoise, … RDA FAIR4RS WG. (2022). FAIR Principles for Research Software (1.0). [https://doi.org/10.15497/RDA00068](https://doi.org/10.15497/RDA00068)
- Ding, K., Zhou, M., Wang, H. et al. A Large-scale Synthetic Pathological Dataset for Deep Learning-enabled Segmentation of Breast Cancer. Sci Data 10, 231 (2023). [https://doi.org/10.1038/s41597-023-02125-y](https://doi.org/10.1038/s41597-023-02125-y)
- Evelina Gabasova. (2016). Star Wars social network (1.0.1) [Data set]. Zenodo. [https://doi.org/10.5281/zenodo.1411479](https://doi.org/10.5281/zenodo.1411479).
- Hansen et al., (2023). TextDescriptives: A Python package for calculating a large variety of metrics from text. Journal of Open Source Software, 8(84), 5153, [https://doi.org/10.21105/joss.05153](https://doi.org/10.21105/joss.05153).

## References (continued){.smaller}
- Lin, D., Crabtree, J., Dillo, I. et al. The TRUST Principles for digital repositories. Sci Data 7, 144 (2020). [https://doi.org/10.1038/s41597-020-0486-7](https://doi.org/10.1038/s41597-020-0486-7).
- Silvester, N., B. Alako, C. Amid, et al. (2015). Content discovery and retrieval services at the European Nucleotide Archive. Vol. 43 , pp. D23-D29. [DOI: 10.1093/nar/gku1129](https://doi.org/10.1093/nar/gku1129).
- The Turing Way Community, & Scriberia. (2023). Illustrations from The Turing Way: Shared under CC-BY 4.0 for reuse. Zenodo. [https://doi.org/10.5281/zenodo.7587336](https://doi.org/10.5281/zenodo.7587336).
- Wilkinson, M., Dumontier, M., Aalbersberg, I. et al. The FAIR Guiding Principles for scientific data management and stewardship. Sci Data 3, 160018 (2016). [https://doi.org/10.1038/sdata.2016.18](https://doi.org/10.1038/sdata.2016.18)