IDBS Data Management Blog

Part two of our ‘smart lab’ blog series around ‘getting the data right’ looks at managing external partner relationships and standardization challenges, based on the infographic released by Smart Lab Exchange (#SLABx).

Managing Relationships with External Partners

Scientific relationships, like social ones, are really all about communication. People used to ‘talk’ through the crafted written word, now largely replaced by instant communication through high context voice and video. Scientific communication is about and through data, but we need to break out from the idea that document or file-sharing is today’s best answer. We need to make sure scientists are able to securely access each other’s high context data at the right time. This actively stimulates quality discussion. Dropboxes and file shares do not. Using granular security systems can host multiple collaborating parties and allow them to secure high quality data in a consistent way. They can then choose to share some or all of it within the collaboration (#collaboration). This can mirror exactly the collaboration agreement between the parties.

The smart data principals of context and connectivity have enabled telecoms communities to create networks with massive value. R&D communities should learn from this success and build the quality systems, datasets and collaboration tools that will enable their Big Data (#bigdata) to deliver Big Collaboration for Big Science.

Lack of Standardization

John Reynders, Vice President, R&D Information at AstraZeneca, reminded the JP Morgan Big Data audience in January ‘The future is federated’. This reflects the need to deliver connected R&D against a background of distributed data. Most sane CIOs recognize that creating ‘Death Star’ mega warehouses to drive standardization is impractical and if it involves data such as patient records, also potentially unethical. So how do we initiate standardization in a federated world?

Ontologies have a massive role to play in driving simplification – whether they are internally or externally curated – those which plug into process applications are key. They drive how data is captured and contextualized, a smart approach which enables high value data assets to become interoperable and comparable.

Lon Cardon, SVP at GSK said at the same meeting: “we just need the right approach to noisy datasets.”We believe that this is missing the opportunity to learn from other sectors such as the telecoms industry, where effort and investment is focused on reducing the noise and the gaps, not simply accepting them and filtering them out.

No more excuses

R&D data exists across multiple systems, disciplines and locations. So long as this data can be linked, through context and provenance, it can be made use of. Building a strong foundation of quality, contextualized data is the key.

In today’s cloud-enabled world of extendable bandwidth, the old limitations of scale no longer apply. Gartner Inc. recently highlighted the availability of global R&D knowledge management systems that support multidiscipline collaboration. Where enterprise class systems like E-WorkBook exist there is no reason why high quality data capture, contextualization, ontology and security should remain long term strategies. Putting these in place NOW in ’SmartLabs’ around the world will enable truly smart R&D enterprises.

So much gets written about the woes of R&D. It’s time to stop. Think. Act. That’s our message at this year’s IQPC SmartLab’s conference in Munich (#SLABx) this week. Getting the data right inside today’s ‘SmartLab’ enables smarter enterprise R&D tomorrow and in the future.This first of two blogs kicks off today with integrating legacy systems and managing secure data flow, based on the infographic released by Smart Lab Exchange.

R&D (#R&D) creates and uses data assets. Like any manufacturer it needs to understand where its assets are, how good they are and how to put them together to make a quality product.The advantage of data as a product is that it can be used, reused and repurposed again and again.

The smart principles of R&D data are straightforward:

• Capture data with context
• Make sure you understand the provenance of the data
• Structure it so that it can be combined and consumed by decision-makers in their own way

Straightforward to say but to achieve this data manufacturers need to think and act for the long term, not just immediate ROI.

Integrating Legacy Systems

Some legacy systems,in-house or COTS, form parts of a fixed process that does not need to change. These systems act as feeders for a foundation of quality data and should be retained. They can be driven by other systems such as process ELNs (#ELN), and their data harvested for use elsewhere. Integration through RESTful web service APIs is the most flexible approach but where the technology can’t be applied, bespoke integrations can still be of value to source the right data and context.

However, where a legacy system – particularly one of the many legacy in-house solutions – is standing in the way of progress,the smart approach is to retire them fast and get their function replaced by a cross-domain, multi-process application, such as E-WorkBook. There is no future in making a data compromise for expediency if the net result is poor quality data.

Managing Secure Data Flow

In a today’s multiparty R&D environment it is vital that data, process and context are stored securely. But what does that mean? Firstly, it’s about the individual: their profile, group and role should define what they can do, and what they can see.  Secondly, it’s about data provenance: all data should have audit trails from capture, through any modification, all the way to consumption.

This approach is standard practice in regulated environments but the principle applies everywhere, even if you do not need the GxP rigour at the bench. Thinking about security and audit right at the start is smart because it is too hard to retrofit down the line. If you get the security and audit right at the early stages, then whatever workflow and orchestration tools you have can be used to traffic data to the right place and in front of the right person.

My next blog will look at managing relationships with external partners, the lack of standardization and how data quality is key, because contextualized data underpins everything.

Scientific research is changing. Our customers want to break down the silos between departments, improve communication, simplify the way they work and let scientists focus on doing science. Their biggest frustration, – especially in chemical research - is that they know they are repeating work that has been done countless times before. But there’s no way to know in advance that a particular chemical synthesis is doomed to failure. Unless you’re lucky enough to have a colleague who has tried it before, you have no way of finding out.

So you go ahead and run the synthesis. Then, for some unknown reason, it fails. You shrug and  – and this is the important part – you don’t tell anyone about it. And that, in my view, is a huge mistake.

Failure is good

Of course, you record the fact that the experiment was performed, exactly as you should, in your paper or electronic journal. Maybe someday one of your colleagues, trying to do the same thing will see your previous attempts, and learn from them. But scientists are cautious about publishing failure. Where is the benefit? Journals aren’t interested in science that doesn’t work. Funding bodies are unlikely to reward someone with a track record of failure. So this huge corpus of potential knowledge is lost. Some sources say only 5% of scientific work ever gets published, the rest of it (the less successful or more humdrum work) sits in paper or electronic repositories, never to see the light of day again.

However, people are beginning to recognise that hiding less than perfect results is a problem, so we, along with many others, are trying to address it.

Instant messaging – for chemists

The advent of free online chemistry resources like the RSC ChemSpider.com, now makes it easy for researchers to find information on compounds. Recent work done by the RSC, in collaboration with the Unilever Centre for Molecular Informatics at Cambridge University, allows researchers using our ELN, E-WorkBook, to look up chemical structures on Chemspider. Here you can see physicochemical properties, assay results and commercial availability for example, all from within your Notebook, by simply pressing a button. And you can share your knowledge more freely, by publishing your compounds to Chemspider. It only takes one click.

Work is well advanced on a way to publish synthetic reaction schemes too, so even those failed reactions that are no good to anyone (except, of course, they are) can go in there too.

I think this is a great thing. It lets chemists learn from each other’s mistakes. It makes more scientific research public knowledge. It breaks down barriers. And it lets chemists communicate their findings with each other openly – hopefully to the betterment of all researchers everywhere. Even better it’s now available free for our ELN users. Take a look at it in action here.

If you want to share your research data with the scientific community – whether it’s a breakthrough or a letdown – or you just want to find out fast information on compounds, download the plugin from our IDBS Labs support portal today.

The more you open up, the more you’ll be amazed at what you can learn.

Say what you’re not saying, don’t say it, say what you didn’t say

Last time I blogged that modelling is not limited to software engineering, play and simulation; but is universal in human endeavour. I mentioned that considering accuracy is important but not sufficient in assessing a model. What other considerations are there?

My favourite lens for looking at a model is abstraction. In philosophical terminology, abstraction is about grouping concepts together at decreasing levels of detail. So, a duck is a duck and no other thing is a duck (no matter how it looks or walks or sounds); but applying abstraction allows us to talk about birds and say useful things, which might be rather exasperating if we had to list every bird in the world to say them. This kind of classification is a particular feature of object-oriented programming languages (which may or may not be a good thing).

Leaving it out

A modeller, not saying

However another way of considering abstraction is to pause before asking what a model is saying, and ask: what is this model not saying?

The model of biological change that we call evolution has incredible empirical support, so that its application has great explanatory and predictive power (some would even say that we don’t apply it enough). Strangely, though, it seems to cause an awful lot of consternation to those who subscribe to another model called creationism.

Why strange? At first sight, both of these models deal with how the world came to be the way it is. But evolution models a process, and has nothing whatsoever to say about how that process began, or why it began, or who began it. Conversely, creationism says nothing about how its proposed agent went about his craft (well, usually). He just did it. Apples and oranges.

Putting it back in

Any critical analysis or use of a model has to carefully stick to assessing or building upon what it actually models. This might sound simple, but humans find it remarkably tricky. We are fond of making cultural and doctrinal assumptions and applying intuitions without knowing about it. (In a black alley, a black cat spies a black rat. How?*)  Unfortunately this is not only inevitable, it’s usually necessary.

Why so? Models almost always rely on background information. Of particular interest in computer science and artificial intelligence is the notion of semantics: the meaning of symbols. Tell a robot to fetch you a cuppa, and it may suffer the same semantic confusion as is now affecting US readers: a cuppa what?

IDBS’ Enterprise Customers range from global enterprises to academic research institutions.  When we work with our customers to document their implementation experiences we hear a similar story, regardless of the industry sector or size of the organization. They all expected IDBS to be a cost effective, robust and secure R&D data management solution, but they also cite the following as additional benefits accruing to their decision to select IDBS over other software providers:

• Significant time savings and increased staff productivity
• Simplification of workflows
• Enhanced tools for collaboration both internally and externally
• Flexible framework & rich functionality
• Process improvements across diverse disciplines
• Domain expertise of the IDBS Support and Professional Services staff

At recent IDBS Roadshows and User Group meetings we have heard from a range of organizations using IDBS, including Solae, Amgen and Abbott Laboratories. Sharing these experiences (the highs and lows, best practices, lessons learnt and customizations) are invaluable for the IDBS community, and also for those organizations that are just starting to work on their own projects using IDBS. A selection of IDBS customer announcements published recently includes:

• PharmaLegacy Laboratories – achieve a 10x improvement in validated reporting time
• Ablynx – deploy IDBS to speed antibody research and discovery
• BASF – use IDBS as its enterprise research data management platform
• Lonza – adopt IDBS to optimize bioprocess execution and knowledge management
• AnaptysBio – deploy E-WorkBook to manage data
• University of Nottingham – extends E-WorkBook Suite across global network of campuses

A full list of case studies can be found on our web site for you to browse. If you want to share your own experiences of using IDBS we’d like to hear from you.