Category Archives: SciDataCon 2016

Posts relating to SciDataCon 2016, held in Denver, Colorado as part of International Data Week, 11-17 September 2016

Humans of Data 9

IMG_3795-edit copy

“We need more south–south collaborations.  I’d like to approach this and get in touch with people I’ve met here, and I’m trying to identify other people in Latin America that have the same interests.  Our data problems might be different from England or Canada or elsewhere in the north.  We have a lot of data that might be at risk of disappearing in the next few years, and this might be a bigger problem in developing countries.

I’m also concerned about how the southern hemisphere is going to contribute.  How do I get the funds that I need to get the work done that I need to do?  Trying to be part of this community is going to be a challenge for financial reasons.  I would surely not be here except for GEO and CODATA support; this was very special for me to receive that funding.  Otherwise I would miss this incredible opportunity for networking and knowledge sharing.

I think that open science is the only way forward to answer the complex problems that have been presented by society.  These problems are not local and involve so many different knowledge domains.  We need to do science from a more collaborative perspective to be able to tackle these challenges.  Collaboration is what I’m really passionate about.  When I return to Brazil I’ll start to talk to people and see how we can go from here.”

Humans of Data 7

img_3843-edit
“I entered into the data profession about three and a half years ago. I found the community to be very welcoming. The ideas of ethics and sustainability are starting to be brought forward more strongly now. Data aren’t just digits in the memory. They have real world effects in real world situations.

One of the things that drew me particularly to the idea of preserving data, is to build on the research investments that people have made. People spend their lives exploring questions. If the information and data those answers are based on, aren’t kept useable in an understandable way, then the answers themselves are also lost. The end result is so many wasted lives when you add it up. It’s the time invested in the exploring these questions, but even more in a broadly humanitarian way, these answers are pursued to improve the lot of humanity. If the data collected through research are lost, the answers themselves are lost, and so the people, the environmental effects are also lost. So I think that’s my most important concern.

Look, I like efficiency. I like effectiveness. Not taking care of things you’ve spent time making, not making sure they can be used effectively – that’s a waste of everyone’s time and effort. It just bugs me. Data is the starting point for any answers we achieve through research. Let’s not waste that effort. If there’s anything this community could respond more in, it’s the human-related areas – the marketing and advertising of the importance of data and the importance of making sure the data is there to go back to. There’s no reason to reinvent wheels, but improving them is vital.”

HarassMap at SciDataCon on their Data Management Project with IDRC

This post comes from Reem Wael, Director of HarassMap http://harassmap.org/en/: Reem was assisted in part by CODATA and GEO to attend SciDataCon and contributed a paper to a session on ‘Data Sharing in a Development Context: The experience of the IDRC Data Sharing Pilot’ http://www.scidatacon.org/2016/sessions/56/

hm_logo_05HarassMap launched five years ago with the mission to end social acceptability of sexual harassment in Egypt. This mission, unexpectedly, led to the accumulation of a lot of data coming from both online and offline sources, and the more we grow the more data we have. Our methodology is to combine online and offline work to achieve our mission and therefore we crowdsource reports on sexual harassment and through our social media outlets, and we receive information from outreach activities and trainings. We analyze this information and give it back to the community in the form of research reports, public campaigns, trainings and policies.

A few years ago, we started receiving requests from researchers who are working on topics that cross-cut with sexual harassment, to access our data. We responded to these requests by providing an excel sheet with the downloaded crowdsourced reports, but this was the limit of our assistance. When an opportunity came along to design and implement a data management plan, supported by IDRC, it was very relevant to our needs. IDRC is an international research organization and was interested in exploring how grantees can have their data more open to the public.

The main point that IDRC focused on is openly sharing the data. However, when we started to work on the project, we realized that the earlier stages are more challenging; which data do we store and how? We have a massive amount of data accumulating in the last five years. Other than the crowdsourced reports and the reports that we receive on social media, we also own a huge library of photographs and video footage, reports from trainings, evaluations from trainings that reflect the impact that we had, reports from outreach activities, and social media posts and replies. We reached some decisions in the planning phase and we are continuing to make these decisions as we move on.

We formed a ‘data management team’ from HarassMap staff who works on research and data and we tried to identify the data that we want to collect, organize and share, raising the following questions: why are we sharing data, and with whom? How can we organize it in a way that would be helpful to researchers, or others who request access to the data? Are there any ethical issues that we need to consider while sharing the data? These questions brought up some challenges. We were not sure what kind of data would be interesting to researchers, for instance. We found that even though crowdsourced reports are more coveted by researchers the more interesting data is the discussions on social media (our posts, including all the comments that we get) in addition to field reports. This data mirrors and tracks the development of myths and misconceptions of sexual harassment, especially when analyzed over a long span of time as it can show if a difference in attitudes and opinions on sexual harassment had occurred.

Embarking upon data management showed some challenges as well. One is a linguistic/technical challenge especially with the crowdsourced reports as we receive them in both English and Arabic. Privacy was a challenge regarding HarassMap’s library of photos and videos since it shows a lot of volunteers since 2010 from whom we did not take consent to share their photos publicly. We did not find an ethical problem with publishing and sharing crowdsourced reports because they are all anonymous, and we also filter them to remove any information that can hold us legally liable such as accusations against people or places by name.

That said, we are now in the process of accumulating and organizing data from the last five years, and putting it on our web server. The next phase – sharing – has its share of challenges. The first and most important is that we must have some kind of screening over who uses our data for various reasons; sometimes researchers completely misuse the data which puts HarassMap in a bad position. For instance claiming that crowdsourced reports reflect ‘hotspots’ of sexual harassment is essentially flawed yet a widely used claim. We always assert that crowdsourced reports provide biased data because there is a huge difference to access to internet and technology based on the affluence of the area and therefore receiving reports from a specific area doesn’t necessarily mean that harassment is more prevalent there, it may mean that people have better access and more knowledge about reporting. At other times, researchers have taken the data without giving credit to HarassMap; and some researchers have asked for the data and then disappeared without informing us of what they wrote.

Being part of this project has benefited HarassMap greatly not only because we started thinking about the idea of sharing our data on searchable engine, but also because we did not know the amount of data that we possess in the first place until we started looking for it. While putting our data completely public is something that HarassMap is still hesitant to do, we are definitely happy to provide researchers and other interested parties with data in a format that is more user friendly.hm

Humans of Data 4

img_3758_small1“So when I was a kid, obviously Star Trek was the thing, because it was our better selves in the 23rd century. Civil rights, women’s rights, all those issues that were happening at that time in the 1960s were simplified in that show. But the thing that got me was the computer. Spock would have this conversation: ‘Computer, what is this thing? What was the global temperature in 1934?’ And there was always an answer. My start with data was looking at how instruments recorded it. As I’ve started to get into managing people, writing code, I’ve realised that we’re the people in someone else’s past. If we don’t get it right, they will suffer. They’ll ask the question, and the computer won’t have an answer. These people are all trying to get to that better 23rd century. It’s slow progress, baby steps. But being able to make sense of the research results that we take now, consolidating that, is really important to me.”

Open Data as a Moving Target: What Does it Take to Allow Reuse?

By Irene Pasquetto

As we all know too well, making all scientific data technically and legally accessible to img_20160913_133322all researchers is an ambitious task complicated by constantly evolving social and technical barriers. It is fair to say that we are making progresses in this direction. At Scidatacon 2016, we examined several concrete solutions that can facilitate openness of scientific data or, if you prefer, make sure data are FAIR (findable, accessible, interoperable and reusable).

However, it seems that the more we learn about how to make data open, the least we know about how exactly data will be reused by the scientific community, which means by the researchers who generated the data and should have a primary interest in accessing it. Very few empirical studies exist on the extent to which open data are used and reused once deposited in open repositories.

The fulcrum of the problem is that data take many forms, and are produced, managed and img_20160913_133754used by diverse communities for different purposes. Nevertheless, different stakeholders (publishers, data curators, digital librarians, funders, scientists etc.) bear competing points of view on the kind of policies, values, and infrastructural solutions necessary to make data open. During a session moderated by Christine Borgman (UCLA) and titled “How, When, and Why are Data Open? Competing Perspectives on Open Data in Science”, Matthew Mayernik (National Center for Atmospheric Research), Parsons Mark (National Snow and Ice Data Center) and Irene Pasquetto (UCLA – Center for Knowledge Infrastructures) presented on some of those challenges that make the use and reuse of “open data” such a complicated and heterogeneous process.

Mayernik argued that the integration of the Internet into research institutions has changed the img_20160913_140145kinds of accountabilities that apply to research data. On one hand, open data policies expect researchers to be accountable for creating data and metadata that support data sharing and reuse in a broad sense, in many cases, to any possible digital user in the world. On the other hand, providing accounts of data practices that satisfy every possible user is in most cases impossible.

In his talk, Parsons effectively showed that data access is an ongoing process, not a one-time img_20160913_134549event. Parsons and his team examined how the data repositories products and their curation have evolved over time in response to environmental events and increasing scientific and public demand over several decades. The products have evolved in conjunction with the needs of a changing and expanding designated user community. In other words, Parsons’ case study shows that it is difficult to predict the users of a data service because new and unexpected audiences (with specific needs) could emerge at any time. Parsons also argued that, for this reason, “data generators” may not be the best individuals to predict future uses of their own data.

Because open data users change over time, it is also necessary to built open repositories that provide data in formats flexible enough to allow different approaches to data analysis and integration, for different audiences. This was the point made by Pasquetto, whose case study is a consortium for data sharing in craniofacial research, with a focus on the subfield of developmental/evolutionary biology that recently adopted genomics approaches to knowledge discovery. Pasquetto found flexible data integration to be a necessary precursor to using and reusing data. “Data integration work” is the most contested and problematic task faced by the community, where data need to be integrated at two or more levels and these levels require extensive collaboration between engineers, biologists, and bioinformaticians.

Borgman also presented a paper on the beneath of Ashley Sands, who recently graduated img_20160913_135716from the department of Information Studies at UCLA and is now senior program officer at the Institute of Museum and Library Services in Washington DC. This talk examined characteristics of openness in the collection, dissemination, and reuse of data in two astronomy sky survey case studies: the Sloan Digital Sky Survey (SDSS) and the Large Synoptic Survey Telescope (LSST). Discussion included how the SDSS and LSST data, and datasets derived from the projects by end users, become available for reuse. Sands found that the rate at which data are released, the populations to which the data are made open, the length of time data creators plan to make the data available, the scale at which these endeavors take place, and the stages of these two projects all have great impact to the extent in which data and then reused.

Moral of the story: open data is a fast moving target. In order to enable reuse, data repositories better start to run.

Humans of Data 3

img_3715“I find it relaxing to work with data.  I’m a mathematician by training and much more into applied mathematics, so I find recursive formulas very relaxing and linear algebra is like a fun puzzle, like a crossword.  I like problem solving.  ‘Big data’ is an excellent field for problem solving.  I like finding elegant solutions to complex problems.  I approach problem solving slightly off-kilter from others – I would often get weird grades in school, but it also means that if people give me problems they’re struggling with, I could look at it and come up with something different from them.  This is my first data science meeting.  I’m enjoying the opportunity and being around mathematicians and database people and folks who get excited by data.  And I’m pleased that there are other women I can talk to.”

Humans of Data 2

img_3635-copy“One of the coolest thing is starting out as a student in the research data management field, being early in my career, and then being able to interact with the same people over time. I feel like I’m kind of growing up as an individual. I feel I can say, hey, you guys made an impact on what I do, and now I can give back.”

Humans of Data 1

img_3656_small“I think you need to express yourself the way you feel you should, because what really matters at this conference is that we’re all interested in making data available, accessible and preserving it, and we shouldn’t feel that we have to sacrifice who we are in part or whole, in order to do our work.

I hear far more people who are complimentary about the way I dress than not, so it’s not like it’s problematic. But it shouldn’t matter anyway. We have to just keep being who we are, and the other people will catch up.”

Scidatacon: Opening keynotes

It was a pleasure to start off the first full day of SciDataCon with a keynote from Elaine M Faustman, Professor and Director at the Institute for Risk Analysis and Risk Communication, University of Washington and member of the ICSU World Data System Scientific Committee.  Professor Faustman’s keynote talk, ‘Challenges and Opportunities with Citizen Science:  How a decade of opening1experiences have shaped our forward paths’, introduced a welcome early focus on the importance of rigorous ethical approaches to ‘citizen science’ research projects. Looking back to the early roots of the knowledge practices we now call ‘science’, Faustman reminded us that of course the roots of these practices can be found in the work of European gentleman scientists and their cabinets of curiosities.   She also situated contemporary citizen science practice in the US legislatory framework of the US citizen’s right to know, work which has been underpinned by standards and acts since the 1940s onwards.

Reflecting over a decade of citizen science practice in the environment and public health domains, Faustman provided examples of projects where citizens are not only research subjects but are centrally influential in the work, to the point where they express ownership of the project alongside the university team.  Discussion focused on the importance of the abilty of research participants to influence the direction and scope of the research project, to provide feedback on its progress, and to have access to the data accrued in order to be able – in case of public health projects at least – to use it to guide their ownopening4 decision-making.  The message of deep ethical engagement and building respectful relationships with participants set the scene for a day in which ethical issues reverberated.

The second keynote was by Simon Cox, Research Scientist, Environmental Informatics, CSIRO Land and Water, Clayton, Melbourne, Australia.  In his talk, ‘What does that symbol mean? – controlled vocabularies and vocabulary services’, Cox raised a very pragmatic point about the widespread problem of non-systematic use of symbols – and keywords – in data.  He demonstrated that we assume symbols and keywords have some sort of shared meaning, at least in a given community, but that the reality is much less systematic. Symbols and abbreviations with no widely used consistent meaning are often used by researchers when creating data. Populaopening6r terms describing volume can mean entirely different things in different countries.  And even symbols of terms describing a widely understood measurement, such as the metre, can be problematic link to a common source: the International Bureau of Weights and Measures provides a definition, which can be found via a given URI. But the fact that this URI has changed regularly from year to year disrupts any expectation of a stable, enduring location for this definition.

Cox suggested a couple of actions to mitigate this situation. Firstly, a new CODATA task group on coordinating data standards will take this work forward. Secondly, the Global Agricultural Concept Scheme – GACS – is the result of three defining sources from agricultural research banding together to deduplicate their respective vocabularies and make them interoperable for agricultural researchers. Cox noted that the technical job is not large but that – in confluence with Faustman’s earlier message – the really big job is achieving the buy-in from the community in question.

So the pesky human dimension appears right at the start of International Data Week!  More information on the keynotes is at http://www.scidatacon.org/site/opening-keynote/

Laura Molloy is a doctoral researcher at the Oxford Internet Institute and the Ruskin School of Art, University of Oxford. She is on Twitter at @LM_HATII.

How to Address Data Challenges in the Biomedical field: Solutions for Data Access, Sharing and Reuse

Irene Pasquetto is a PhD Student in Information Studies at UCLA.

As scientists in the biomedical fields are generating more and more diverse data, the real elaine-m-faustmanquestion today is not only how to make data “sharable” or “open”, but also, and especially, useful and reusable. At Scidatacon 2016, speakers from funding agencies, research universities, data research institutions, and the publishing industry came together to try to address this key question.

Around 20 highly interdisciplinary papers organized in four busy sessions addressed the problem from different perspectives, while agreeing on an essential point: developing new, open frameworks and guidelines is not enough. Indeed, what characterized this last edition of Scidatacon was a focus on proposing and discussing applicable solutions that can address the management, use, and reuse of large scale datasets in biomedicine today, right now.

Three main themes emerged across the sessions:img_20160912_125708

  1. How to enable scientific reproducibility.
  2. How to apply data science techniques to biological research.
  3. How to make heterogeneous bio-databases globally interoperable.

#1 HOW TO ENABLE SCIENTIFIC REPRODUCIBILITY

Leslie McIntosh (Director, Center for Biomedical Informatics Washington University in St. Louis) moderated session 1, which focused on the first topic: Solving the problem of reproducibility in science, starting from making jennie-larkin-biomedical-data-stewardshipbiomedical data reusable to this end.

Tim Errington (Center for Open Science) offered a clear and useful distinction between reproducibility, which he defined as the possibility of re-running the experiment the way it was originally conducted, and replicability, which is the possibility of getting the same results by reusing the same methods of data collection and analysis with novel data. Errington invited the audience to reflect on two main issues: first, incentives for individual success are focused on “getting it  published, not getting it right,” and second, instead of focusing on problems with either open access or open data, we should think about “open workflows” that include the whole process of scientific research.

Similarly, Anthony Juehne (Washington University in St. Louis) talked about how to address reproducibility issues step by step across the entire “scientific workflow”. Juehne presented to possible solution to the problem: “Wrap, Link, and Cite” data products OR “Contain and Visualize” them using virtual machines.

Finally, Cynthia Hudson Vitale exposed a rarely addressed aspect in the reproducibility community, which is the fundamental role played by biocurators. While their work is often not acknowledged in the community, biocurators are those who de-facto do the hard job of cleaning and organizing the data in a way that can be used to reproduce experiments. Cynthia proposed some concrete solutions to the problem. First, domain reproducibility articles need to include a greater variety of curation treatments. And, second, curators need to publish in domain journals to ensure the full breadth of curation treatments is discussed with researchers.

#2 HOW TO APPLY DATA SCIENCE TECHNIQUES TO BIOLOGY RESEARCH

A second main theme that emerged in session 2 was how to apply recent statistical and jiawei-han-large-scale-biological-text-mining-and-data-analysis jiawei-han-panel-large-scale-biological-text-mining-and-data-analysiscomputational cutting-edge techniques for data science (machine learning algorithms, deep learning text mining) to the biomedical knowledge discovery process. Introduced and moderated by Jiawei Han (University of Illinois at Urbana-Champaign), computer scientists, biologists and biomedical researchers working on biological text mining presented overviews and surveys on the topic.

Beth Sydney Linas and Wendy Nilsen from IIS, the Division of Information and Intelligent Systems (NSF – National Cancer Moonshot), gave an overview of how data science can be used to uncover the underlying mechanisms that drive cancer and the development of methods that will allow clinical researchers to eliminate the disease. The researchers concluded that the future of novel computing (especially machine learning, artificial intelligence, network analysis, database mining as well as bioinformatics and image analysis) needs to be directed also as it relates to health related research.

Elaine M. Faustman (University of Washington) presented an annotated database of DNA and protein sequences derived from environmental sequences showing AR in laboratory experiments. The database aims to help fulfill the current lack of knowledge on the relations between antibiotics resistant genes present in the environment and genomic sequences derived from clinical antibiotic resistant isolates.

Jiawei Han, Heng Ji, Peipei Ping, Wei Wang presented results from their analysis of massive collection of biomedical texts from medical research literature using semi-supervised text mining. The researchers argued that interesting biological entities and relationships that are currently “lost” in unstructured data can be efficiently re-discovered by applying bio-text mining techniques to PubMed massive biological text corpus.

# 3 HOW TO MAKE HETEROGENEOUS BIO-DATABASES GLOBALLY INTEROPERABLE

Finally, over 10 presenters in session 3 and 4 shared their own first hand experiences in susanna-assunta-sansone-biomedical-data-stewardshipmanaging and building biomedical integrated databases and making them interoperable. The biomedical research community and funders seek to make their research resources “FAIR”: findable, accessible, interoperable, and reusable, and also seek to strengthen incentives to support improved data stewardship by addressing incentives, such as data citation. Speakers shared a common concern: how to create data standards and practices from the bottom-up. As suggested by the speakers, it is necessary to be aware of existing local, cultural and social incentives, clearly define possible audiences, and involve the scientists in the database-building process. Individual projects can be consulted at the sessions’ webpage: http://www.scidatacon.org/2016/sessions/34/