The Integrated Lab » Academia

Semantic annotation of experimental records

John Trigg — Fri, 30 Sep 2011 11:02:14 +0000

A recent article ‘Where do the ideas come from?‘ I had written for Pharma IQ kicked off a short discussion on a LinkedIn group (LIMS Forum) that included a reference to a paper on ‘Annotating Experimental Records using Ontologies’ by Alexander Garcia, Olga Giraldo, and Leyla Garcia from the University of Arkansas for Medical Sciences, Biomedical Informatics, Arkansas, USA. The paper addresses a growing concern within the laboratory information and knowledge management domain about the ability to efficiently and effectively retrieve records from Electronic Laboratory Notebooks. Their approach was to adopt sematic annotation by combining existing ontologies that support organisation and classification with user generated tagging to generate a three-layer model that relates to the document, the annotation and the experiment. The model allows complex queries based on SPARQL. At present the project is at the prototype stage, but the continuing effort includes plans to release the software to the open source community.

The Smart Laboratory: Towards a National ELN

John Trigg — Mon, 01 Aug 2011 10:11:57 +0000

This was a 2 day meeting, as part of the Lab of the Future theme of the EPSRC Dial-a-Molecule Grand Challenge, held in Southampton, UK on 28-29 July.

Dial-a-Molecule is funded by the Engineering and Physical Sciences Research Council (EPSRC), the main UK government agency for funding research and training in engineering and the physical sciences, investing more than £850 million a year in a broad range of subjects – from mathematics to materials science, and from information technology to structural engineering.

Central to Dial-a-Molecule is to be able to predict reaction outcomes, and the availability of high quality data on all reactions carried out is a key enabling resource. A Smart Laboratory must have the collection and use of data at its heart, and a ‘National’ Electronic Laboratory Notebook is a potential means of achieving this objective.

The meeting was attended by a total of 37 representatives of academia, industry and ELN vendors. The programme was a mixture of presentations and workshop sessions. The presentations concentrated on examples of successful deployment of ELNs in industry, the work being undertaken by the Pistoia Alliance to develop standard approaches to data sharing, and ELN initiatives already underway in some academic institutions. In addition, vendors were given the opportunity to present strategies and key features of their products.

The workshop sessions addressed a number of topics
• Why has uptake of ELNs in academia been poor?
• What features would make adoption compelling?
• What other features would we like?
• Any problems that would limit adoption?
• Planning the deployment of an ELN system. Cost estimates.
• Establishing and using a national compound registration system.
• National ELN – beyond synthetic chemistry.
• Defining a common format for data exchange

The outcome of the meeting will be formally reported through appropriate channels in due course, but here are a few personal observations.

• The underlying problems that an ELN can address in industry and academia are very similar, but the context is somewhat different. Primarily it is an issue about intellectual property and the ability to productively manage, use and share IP over its life cycle. The context varies in two ways; firstly, funding and maintaining an ELN in academia presents a different type of challenge. Laboratory productivity will be seen as more of a critical activity in industry (discovery, development, measurement) and will therefore be more readily identified as an area for process improvement. Secondly, laboratories in academia are historically more tolerant of individuality in the choice and use of tools for documenting experiments. The consequence of these two points is that the need for an ELN in academia will require a bottom-up approach with lots of inertia, rather than the top-down that is more common in industry.

• A presentation by Richard Bolton (GlaxoSmithKline) on the progress being made by the Pistoia Alliance ELN Query Service Project in developing a standard approach to sharing ELN data showed considerable consistency with the objectives of the data sharing requirements of the National ELN initiative. If the assumption is that ELN vendors will, in due course, sign up to the Pistoia query standard as it evolves, then a National ELN can take full advantage.

• There are a number of existing deployments of commercial ELNs in academia, as well as non-commercial initiatives (e.g. LabTrove, the Univ. of Southampton project). Adoption is mixed, and a number of speakers from academia highlighted the issue about take-up, e.g. adoption is 90% a ‘user’ issue.

The output from the workshop sessions will be used to refine ideas about, and requirements for a National ELN. Not surprisingly, this output was reasonably consistent with typical requirements from industry-based project dealing with small molecule chemistry. I’ll watch the next steps of this initiative with great interest.

Semantics and ELNs

John Trigg — Fri, 29 Apr 2011 10:11:12 +0000

I recently conducted an interview on behalf of Pharma IQ with Jeremy Frey, Professor of Chemistry at the University of Southampton on the subject of semantics and electronic lab notebooks. You can listen to the interview here.

Paper vs Electronic

John Trigg — Mon, 08 Mar 2010 15:55:59 +0000

Following on from my post about Notebook Sharing, here’s another blog post wrestling with the consequences of changing from paper to electronic lab notebooks. Interestingly, the prospect of using an Apple iPad as an ELN gets some consideration. I suspect that we’re going to hear a lot more about iPads in due course.

Notebook sharing

John Trigg — Thu, 04 Mar 2010 09:45:17 +0000

One of the big arguments in favour of an Electronic Lab Notebook is of course the ability to share information with your co-workers. However, despite the obvious benefits of sharing, there’s always the ‘cultural’ issue to deal with as well, i.e. how happy am I to share my ideas, records and data with my co-workers? We spend a lot of time worrying about these issues and evolving strategies to deal with them, so it was quite interesting to come across a real life story (Jenski, PhD) that seems to encapsulate the dilemma, if not the solution.

The GPHY* School

Gloria Metrick — Wed, 17 Feb 2010 15:14:47 +0000

It sounds as if there are several schools around the world to train future bioinformaticians and laboratory informatics professionals. It is only on the European projects I have been working on where I seem to run into graduates of such a program, and it always seems the person comes from the GPHY School in Poitiers, France ( http://www-gphy.univ-poitiers.fr ).

The graduates of this school talk enough about it and seem so tightly-linked with it, that I have finally become curious-enough to learn more about it. I figured I would share the information with all of you, at the same time. To learn more about the school, I spoke with Patrick Girard, who has been the school’s Director for ten years, and Claude Goldszmidt, GPHY Professor and Senior Consultant at Capgemini.

The GPHY School was founded in 1971. Although it was initially more geared toward engineering, it was always intended to provide a multidisciplinary curriculum, combining courses such as biology and computer science. The graduates are formed more as computer scientists, but with a strong scientific background so that they will specifically understand the issues of scientific software. The school was initially focused mainly on pharmaceuticals, biotechs and biopharms, but in 2000, it diversified to service an increased variety of other industries that are biology and health related.

90% of the students that come to GPHY have a background in Biology. The school’s success in placing their graduates comes partly from their hands-on approach to learning the discipline. Students are trained in real-life systems using professional methods, and are required to mix internships with companies along with their coursework. The students are required to spend three months of each year of study in internships for their first two years, and six months of internships for their third year of study.

Additionally, the school maintains a strong relationship both with the end-user companies, as well as the scientific software companies. In fact, many instructors are experts from the industry, who share their knowledge and first-hand expertise with these students. The school maintains links with over 500 French companies, and links with companies from other countries, as well.

Students are increasingly hired by both these categories of companies, as the school maintains ongoing relationships with these companies in order to understand what job skills are needed. Thus, the school maintains that their students, upon graduation, will have a shorter ramp-up time than students who did not receive this type of multi-disciplinary and work-study-type of training. It is the school’s goal to provide as much real-world experience and knowledge as possible to prepare their students to excel in their real-world positions.

Thank you to both Patrick and Claude for the time they gave to me to help me put this information together to share with all of you who read TheIntegratedLab.com.

* GPHY means: Genie Physiologics Informatics.

In praise of the lab notebook

John Trigg — Mon, 03 Aug 2009 08:20:26 +0000

Elizabeth Moritz started an interesting post on Nature Network entitled I Want YOU…to write in your lab notebook. Interestingly, the post is not about technology, just the discipline of keeping a lab notebook. The comments are particularly revealing, and well worth reading!

Scientific Library Social Networking

Gloria Metrick — Mon, 27 Jul 2009 11:44:39 +0000

I just finished reading “Social Software in Libraries: Building Collaboration, Communication, and Community Online” by Meredith G. Farkas. I was reading it for tips on managing my LIMS/Laboratory Informatics group, which I figured is also a place to provide content and share ideas, in some small way, a little like what a library is meant for.

And, by the way, if you get this book and skim it, you can pick up some good tips for starting blogs and other social networking tools for your company or other purposes, too.

In any case, the book mentioned this Science library blog at Binghamton University as an example of how a library can do a blog and I wanted to pass it along in case it was a good example for anything someone might want to do within their own company:

http://library.lib.binghamton.edu/mt/science/

Academia, industry, biology & chemistry

John Trigg — Mon, 13 Jul 2009 15:49:16 +0000

Following on from some previous posts about ELNs in academia, I came across this; ‘Thinking about electronic lab books’. It’s another creative approach to implementing an ELN, using a wiki, based on five very clear requirements:

Easy access from any computer with a web browser
Ability to securely edit the pages, and keep an edit history
Commenting from other people
Backup
Search

Simple requirments; simple solution.

But on a different note, the University of Cambridge has taken a commercial route. I can’t help thinking that one approach, from the complex area of bioinformatics, has adopted a very simple solution, whereas the other, from a more structured chemistry-based process, has chosen a sophisticated commercial solution. It does draw attention to the difference in requirements between biology and chemistry when choosing an ELN, and the difference in the possible solutions.

The Digital Divide between Scientists

John Trigg — Mon, 22 Jun 2009 08:53:16 +0000

Science always was, and always will be, a subject that promotes and provokes debate. The quest for knowledge and the evaluation of ideas and theories forms the basis of the subject. It seems then that Web 2.0 tools are a natural fit in our wired world to advance debate, and there is a growing body of evidence to confirm that academia is taking advantage of these tools to extend the debate to a wider audience. In the commercial world, however, confidentiality, security and IP protection tend to put some restrictions on open debate, so the discussions are more likely to take place behind the firewall.

Nevertheless, as much as technology can provide the tools to support debate and the sharing of scientific knowledge, there remains an underlying concern about the ‘digital divide between scientists‘. This is a phrase used by Martin Fenner in his blog on scientific publishing in the internet age on Nature Network. It’s pretty much the same concern as we have expressed here on The Integrated Lab about using laboratory systems. Martin’s blog makes interesting reading, not only in articulating the problem and suggesting a few potential solutions, but also from the numerous comments that leave me thinking there’s still a very long way to go before we can stop worrying about the technology and get on with the science. So, just how big is the problem in the commercial world, how much damage is it doing, is it improving, or do we have to wait 10-20 years until we are all digital natives?

Conference Alternatives

Gloria Metrick — Wed, 17 Jun 2009 11:58:42 +0000

In a previous post, I told you why you need to attend conferences in-person and why they provide more value than other events where you do not meet people face-to-face. Unfortunately, some of you don’t have a choice in this. It’s not just today’s economy but we’ve been seeing for quite some of the past years that training and conference budgets have been cut for many people. It’s just that it’s probably even worse when the economy slumps.

And so, let’s consider some alternatives:

1. Webinars: These can be especially good for single topics. If they are run properly, you will get appropriate opportunities to ask questions (by typing and/or speaking). If the people running it do not know how to use the on-line tools, properly, it often ends up as a waste of time to attend, but the well-run ones can be quite good.

2. On-line courses: My comments are the same as for webinars. If the trainer doesn’t understand the tools and the difference between providing a live training course, it will be a problem. Try to avoid taking courses where some people are seeing it live and some are on-line because instructors often focus on the questions of the people sitting right in front of them over those posting questions on-line or via the telephone.

3. Web-sites, discussion groups, on-line social networking: There are a LOT of these available. Most of them have little to no activity. Some of them do have activity but not useful activity. What I mean is this – seeing 100 people wish each other a “happy holiday” is not useful. Being entirely inundated, these days, you can aimlessly look-around the web until you find the right tool. However, you will help yourself in this if you ask others for help. Some of the people you will network with on-line or in-person can point you to sites and tools that they’ve found useful.

Let me point out as I always try to remember to do that none of this is as powerful as going and meeting other living, breathing human beings. Face-to-face interaction brings better results and often provides unexpected and useful information. But these tools are productive for their own purposes and have their uses.

Google Wave and the Laboratory

John Trigg — Tue, 09 Jun 2009 13:50:25 +0000

The impact of new and changing technologies may take some while to filter into industrial laboratories. Concerns about regulatory compliance, IP protection, long-term record keeping and culture will make sure of that. However, there is an undercurrent of interest in Web 2.0 tools (blogs, wikis, etc.) and trying to figure out their value, and what role they could play in a laboratory environment. Additionally, software as a service (SaaS) has suddenly become a popular platform for some LIMS and ELN vendors. Academia, of course, doesn’t have the same level of constraint as industry, and it has been interesting to observe the take-up of emerging technologies and the novel ways in which they may be utilised.

It is now getting to be quite common to find presentations on aspects of Web 2.0 in conferences dedicated to laboratory data management. It was at the SMi Conference on Electronic Lab Notebooks in London earleir this year that Cameron Neylon gave a presentation on ‘The lab that was born on the web’. The presentation was a fascinating insight into the construction of a laboratory system built from available, free of charge, web-based tools. Those of us from an industrial background probably winced a bit, thinking FDA…, USPTO…, and a few other constraining acronyms, but it was quite eye-opening to consider what could be achieved with zero capital outlay.

Cameron is an established blogger (Science in the Open), and he has recently turned his attention to the announcment of Google Wave, a new online communication and collaboration tool, where a “wave is equal parts conversation and document, where people can communicate and work together with richly formatted text, photos, videos, maps, and more.” With a definition like that, you could easily imagine wide potential across a number of industrial sectors. You can access the two parts of Cameron’s thoughts here:
Google Wave in Research: Part 1 – Papers
Google Wave in Research: Part 2 – The Lab Record

ELNs in Academia

John Trigg — Mon, 18 May 2009 09:52:50 +0000

The rise of Electronic Lab Notebooks as an essential tool in the laboratory has occurred in the same time frame as the rise of social network tools for collaboration and sharing outside of the laboratory. Industry is increasingly taking note of the potential of social tools to support ‘knowledge management’ initiatives. At the same time, industrial laboratories are converting from traditional paper lab notebooks to the electronic variety to support scientific collaboration and sharing (scientific knowledge management). Academia, of course, has similar requirements, but without the financial clout (or the inclination?) to purchase and implement fully functional, commercial ELNs. One of the outcomes of this situation is the adoption of social tools in academic labs for experiment write-ups. There’s little doubt that some of the requirements of industry and academia are different, although the gap is probably considerably less than it was some years ago, and with the increasing number of collaborations between the two domains, it is likely to reduce further. So it makes interesting reading to hear some of the innovative approaches to adopting ELNs in the academic world, and to consider whether there may be implications or learnings for industry.

Here’s a few examples: Steve Easterbrook, a professor of computer science at the University of Toronto discusses the use of wiki pages as a notebook, in his Serendipity blog. Dave Lunt, a Lecturer in the Department of Biological Sciences at the University of Hull (UK) talks about the use of WordPress on EvoPhylo. Philip Johnson, a PhD student at the University of Toronto has written about semi-digital lab notebooks, and an earlier post, digital lab notebooks, generated a number of interesting comments. The Department of Neurobiology and Behavior at Cornell University offers a brief overview of matters relating to laboratory notebooks, despite some non-working links. Shawn McLelland from the Neuroscience Department at UC Irvine has a website, e-Lab-book.com (The quest for the electronic lab notebook). Cameron Naylor from the Science and Technology Facilities Council (STFC), the UK’s major provider and supporter of large scale academic research facilities, discusses ‘the integrated lab record or the web native lab notebook’ in his blog on Science in the Open. Also on Science in the Open is a two part article on the Southampton E-Lab Blog Notebook: Part 1 Aims; Part 2 ELN Strategy.