blog

This is the fifth and final instalment of the “What is…” series of posts, in which we focus on open science. Previous posts in this series covered open access, open peer review, post-publication peer review and open data. Here we look at open science as a collective term, and some of the other concepts that fall under the open science term.

Open science is the concept of opening up all aspects of scientific research, to allow others to follow the process and collaborate. There is no formal definition of open science, but it usually incorporates some of the aspects we looked at before, such as open access, open peer review, post-publication peer review, and open data. Additionally, it includes other ways to make science more transparent and accessible during the research process, and we will discuss them here: open notebook science, citizen science, and aspects of open source software and crowdfunded research projects.

 

Open notebook science

While some groups use online, password protected, lab notebooks to share notes with collaborators, open notebook science takes this a step further by making day-to-day lab notes available in real time. By keeping notes online, rather than in an offline lab notebook, open notebook scientists are giving everyone direct insight into their work, and enabling easier collaboration. For example, you can find open notebooks on OpenWetWare (biology and biological engineering), Open Notebook Science Network (chemistry and other disciplines), or The IPython Notebook (interactive computational science).

This is quite a radical form of openness, and few bench scientists use a fully open notebook system at the moment. The general reluctance of many researchers to share ongoing research data is the fear of being scooped by competing groups in academia or industry, as well as being unsure whether they can still publish the work in their journal of choice afterwards. With the increased use of preprints in biology, more journals are developing guidelines about whether they will consider publishing previously shared research, which may alleviate some of the concerns about putting lab notes online.

Open notebook science shares some similarities with open data: both make the underlying research data public. However, where data sharing can occur at the point of publication of the resulting journal article, or after a conference (e.g. by uploading a conference poster or slides), open notebook science happens “live”: data and methods are made public at the moment of collection.

 

Citizen science

One of the most traditional uses of collaborative open research data predates the open science movement: citizen science. Here, members of the public, who are often not scientists themselves, participate in the collection (and sometimes analysis) of scientific data. The oldest running citizen science project is the Audubon Christmas Bird Count, which started in the year 1900. In this bird count, and in similar surveys run subsequently by other organisations (see “further reading” below for a few examples), people are asked to take note of wildlife in their area , and report which animals they encounter. This sort of work helps ecologists survey populations in large areas, and informs long-term conservation studies. Since data are collected by citizens, they are usually made available to the public after analysis. The Audubon counts, for example, are on their website.

In recent years, citizen science projects have moved to the web. Over a million people are registered to participate in Zooniverse’s citizen science projects, which involve anything from hunting for planets to counting penguins. These web-based citizen science projects have a very low threshold for participation, and only require an internet connection and a few minutes of time during coffee breaks. In all cases, they are research projects where human eyes work better than computers, and where researchers need help from a large group of individuals to analyse a large dataset.

Another type of citizen science uses computer games to perform scientific calculations: Foldit is a game that lets players find the best protein-folding conformation, after which human-driven intuitive protein folding solutions are used to optimize computational protein folding calculations.

Such projects fall under open science because the researchers are allowing anyone to interact with their data, but they do restrict and control that analysis to their own platform. In most cases, they will make the resulting publications available to everyone (both Foldit and Zooniverse provide scientific publications via their sites), and in exceptional situations, citizens who participated in these projects may even be listed as co-authors.

 

Crowdfunding

Another feature that is sometimes included under “open science” is the crowdfunding of research projects. It should be noted, though, that crowdfunding does not require open science: researchers can crowdfund closed research projects as well.

However, like citizen science, involvement of a large group of people in the work (now at the funding level rather than at the data collection stage) often encourages researchers to make the output openly available so that donors can have access to the results of the work.

Usually, successfully funded projects have a clear goal, a small budget, and intriguing perks. There are exceptions: one of the most highly crowd-funded science projects is the ARKYD space telescope by Planetary Resources, which raised 1.5 million dollars. Crowdfunding often only works for distinct projects and is difficult to scale to running a lab long-term. That being said, the California-based Perlstein Lab did start out as a crowd-funded project, but is currently run as a seed-funded startup company.

 

Open source software

Open science also overlaps with the open source software movement, which advocates the use and development of software that has its source code made available to others to re-use and build upon.

Open source software for science includes projects like the BioJS library of graphical components, which anyone can use or build on to represent and visualise biological information.

At the moment, a lot of scientific research and communication still relies on software that is not open source, and many open science proponents will, where an equivalent open source alternative is not available, use software that is not open, but will make sure the output of their own work (including software they produced) is openly available.

 

Open drug discovery

Open drug discovery combines open notebook science, citizen science and open source science to find new drugs. Different groups use slightly different approaches, but all are based on open science principles. The Open Source Drug Discovery platform, based in India, uses community participation for initial candidate target discovery, and works with researchers in academia and hospitals to then synthesize and test the targets. Other groups are focused more specifically on one disease, such as the Open Source Malaria project, which uses an open notebook approach to share all ongoing work. Various other collaborations are in place to find and test drug targets for a wide range of diseases.

 

Future of open science

Over the past month we’ve looked at several aspects of open science: open access, open peer review, post-publication peer review, open data, and the topics listed above. That leaves just one final thought: What is the future of open science?

As we’ve seen in previous posts, open access has been growing steadily over the past decade, and open peer review is becoming more popular. Many major funding organisations are asking for not just publications, but also the underlying data to be shared openly, and crowdfunded researchers are also often incentivised to share their work with their funders. Meanwhile, citizen science has been around for over a century, and is only growing with novel web applications that enable everyone to participate in scientific research and drug discovery. Online post-publication peer review is still quite new, and open notebook science has not spread very far yet, but both of these are steadily growing as well.

So despite researchers’ fear of competition, and a reward system that still favours publication in exclusive journals (where openness is not a main concern), scientific research is gradually moving towards an open science system.

We hope this series of posts has been an informative introduction to open science, and would like to invite you to discuss any of your thoughts on Twitter using the #openscience hashtag.

 

References and resources

Open Notebook Science Network https://onsnetwork.org/

Open WetWare: https://openwetware.org

The iPython Notebook https://ipython.org/notebook.html

Audubon Christmas Bird Count https://birds.audubon.org/christmas-bird-count

Audubon Christmas Bird Count data https://netapp.audubon.org/cbcobservation/

Zooniverse https://www.zooniverse.org/ Fold-it https://fold.it/portal/

Foldit publications https://fold.it/portal/info/about#folditpub

ARKYD: A space telescope for everyone. (Kickstarter funded 1 July 2013) https://www.kickstarter.com/projects/arkydforeveryone/arkyd-a-space-telescope-for-everyone-0

How a biotech startup juiced investors with yeast, flies and zebra fish (1 August 2014) https://www.bizjournals.com/sanfrancisco/blog/biotech/2014/08/perlstein-lab-ethan-martin-shkreli-ngly1-biotech.html?page=all

BioJS https://biojs.net/

Corpas M, Jimenez R, Carbon SJ et al. BioJS: an open source standard for biological visualisation – its status in 2014 [v1; ref status: indexed, https://f1000r.es/2yy] F1000Research 2014, 3:55 (doi: 10.12688/f1000research.3-55.v1)

Lab bench image by cjp24 – Licensed under CC BY-SA 3.0 via https://commons.wikimedia.org/wiki/File:Paillasse_chimie.jpg#mediaviewer/File:Paillasse_chimie.jpg

Open Source Drug Discovery https://www.osdd.net/

Open Source Malaria https://opensourcemalaria.org/

 

Further reading:

Center for Open Science https://centerforopenscience.org/

Open Science Federation https://opensciencefederation.com/

Mozilla Science https://mozillascience.org/

Open Science at the Open Knowledge Foundation https://science.okfn.org/

Open Science Framework https://osf.io/4znzp/wiki/home

Science Commons principles for open science https://sciencecommons.org/resources/readingroom/principles-for-open-science/

Woelfle M, Olliaro P, Todd M (2011) Open science is a research accelerator Nature Chemistry 3, 745–748

Helen Shen: Interactive notebooks: Sharing the code (5 November 2014)

https://www.nature.com/news/interactive-notebooks-sharing-the-code-1.16261

The Monarch Program – monitoring monarch butterflies https://www.monarchprogram.org/

Scistarter – database of active citizen science projects https://scistarter.com/

SciFund Challenge – helps researchers communicate more clearly about their work, so that they can be more successful in crowdfunding. https://scifundchallenge.org/

Experiment.com https://experiment.com/

Petridish https://www.petridish.org/

Kraker P., Leony D., Reinhardt W., & Beham G. (2011). The Case for an Open Science in Technology Enhanced Learning. International Journal of Technology Enhanced Learning, 6(3), 643–654.

Sabou M, Bontcheva K, Scharl A (2012) Crowdsourcing Research Opportunities: Lessons from Natural Language Processing. In: 12th International Conference on Knowledge Management and Knowledge Technologies, Special Track Research 2.0 (#STR20) , 5-7 September, Graz, Austria.

MacLean D (2013) Cutting edge: Changing the rules of the game eLife 2013;2:e01294

Li FW, Pryer KM (2014) Crowdfunding the Azolla fern genome project: a grassroots approach, GigaScience 3:16

Masum H, Rao A, Good BM, Todd MH, Edwards AM, et al. (2013) Ten Simple Rules for Cultivating Open Science and Collaborative R&D. PLoS Comput Biol 9(9): e1003244. doi:10.1371/journal.pcbi.1003244