Thanks for visiting Scientific Computing World.

You're trying to access an editorial feature that is only available to logged in, registered users of Scientific Computing World. Registering is completely free, so why not sign up with us?

By registering, as well as being able to browse all content on the site without further interruption, you'll also have the option to receive our magazine (multiple times a year) and our email newsletters.

Life in the cloud

Share this on social media:

In the first of two articles about informatics software as a service, Sophia Ktori looks at the motivations for transferring to the cloud

One of the topical issues in 2015 has been the benefits to R&D-centered industries of transferring their informatics platforms into the cloud and adopting software as a service (SaaS). One of the claimed benefits is the ability to offload to the vendors a proportion of the financial and technical burden of establishing complex and sophisticated software. Talk to the software vendors themselves, however, and there is discussion, if not dispute, and a variety of opinion. According to some, the benefit is purely the financial aspect of not making a large capital outlay to buy an in-house informatics system; others will argue that there are technical benefits in simplifying installation and customisation. A third claim is that the cloud fosters cooperative working, across laboratories and time zones. One thing is certain: the cloud is not one-size-fits-all. 

If you look at ‘the cloud’ as a platform for hosting informatics and data, then what you are effectively referring to is the (off-site) location, and possibly size, of the servers on which your database sits and your application is run, points out John Boother, managing director at UK-based Autoscribe Informatics. ‘There is no massive technological leap. Moving into the cloud is, in effect, a commercial, rather than a technical decision.’

The SaaS license model does provide a way of spreading the cost of an informatics system purchase – rather than requiring up front capital expenditure for a perpetual license – he admited. ‘However, over a number of years, those quarterly, six monthly, or yearly SaaS fees may well add up to the same as the cost of a perpetual licence. What SaaS in the cloud does give you is the potential to reduce the upfront expense associated with onsite IT support, back up, and in-house, or privately rented, server costs.’

Prohibitive capital expenditure

So yes, one of the major advantages of switching over to SaaS in the cloud is financial, agreed J J Medina, senior director of product strategy at GoInformatics. ‘It is increasingly tough for companies to absorb the huge capital expenditure that is inevitably associated with purchasing an on-premises laboratory information management system (LIMS). For a large enterprise, the acquisition of dedicated servers, significant upfront licensing costs, customisation packages, and committing to ongoing maintenance, can amount to a couple of million dollars. This may also involve committing to substantial implementation costs, with a lengthy deployment over multiple years. You end up with standing armies of people on both sides – vendor and client – which will require additional investment from the client to hire and train IT personnel. For most small and medium enterprises (SMEs), in particular, this sort of investment is prohibitive.’

Herein lies a major drawback with implementing on-premises solutions, Medina stressed. ‘Many traditional LIMS packages provide customisation or authoring tools to personalise the system to the company’s requirements. The trouble with customisations is that you have to train dedicated staff to maintain the system, roll them out to additional departments and geographies, and customise for specific application requirements. If and when those staff members move on, the ‘know-how’ goes right out the door with them. The company is then tasked with rehiring and retraining new individuals, or relying on the vendor to provide interim support services at an additional cost.’

There are further financial benefits of cloud solutions that relate to time saved and quality and quantity of usable data collected – in addition to lessening the burden of purchasing, deployment, and maintenance – commented Anthony Uzzo, president and cofounder of Core Informatics. ‘Cloud solutions can effectively and efficiently capture and secure data that has been generated by CROs and collaborators who are outside of the client’s infrastructure. Think about all the money that large pharmaceutical companies spend on outsourcing and collaborative research. If the partners are not all using the same platforms for data collection and sharing, a huge amount of valuable information will likely have to be passed along by email, or via antiquated SharePoint sites, or even just on paper, and so much of this data will lack any scientific context.’

Seamless communication

It is relatively painless and cost-effective for the client and each of its collaborators or service providers to adopt the same cloud platform, Uzzo pointed out. This gives the opportunity to communicate seamlessly, and to capture raw data from the instruments at the CRO or collaborator’s sites. ‘Scientists can then gain maximum value from the use and analysis of that data, and avoid the need to repeat experiments either because they can’t actually find who has carried out the experiment before, or because the protocol, results or data analysis don’t carry any context. This avoidance of repetition saves money in absolute terms, but it also saves time, which can ultimately speed a project through to completion.’

Financial constraints and collaborative R&D models are also ticked off as major drivers for the uptake of cloud solutions by Jens Hoefkens, director of research, strategic marketing at PerkinElmer, and Brian Gilman, strategic marketing, elements at PerkinElmer. ‘There has been an incredible appetite for cloud technology in the life science and healthcare sectors, as companies transfer the costs of maintaining an informatics infrastructure to vendors like ourselves,’ Gilman notes. ‘And with business models for pharma R&D increasingly focused on outsourcing or collaboration, cloud technology makes it so much easier to support data distribution and sharing across a firewall,’ Hoefkens added, mirroring Uzzo’s sentiments.

‘Cloud is also ideally suited to situations and environments in which scale is otherwise an issue,’ suggested Nic Encina, VP, Innovation Lab, at PerkinElmer. ‘Scale may mean housing the wealth of data that is being generated by scientific instrumentation and workflows, or it may refer to computing capacity to enrich and analyse that data as quickly as possible.’

Capex vs opex

Ease of adoption, flexibility and support were additional factors that steered GoInformatics to build its R&D platform in the cloud from inception, Medina explained. ‘The cloud and SaaS business models represent far more attractive options, particularly for the SME sector. SaaS takes less time to set up for each client, and typically requires predictable monthly or annual subscription costs that the company can accommodate into its budget. From an accounting perspective, shifting from making a huge capital expense (capex) to predictable operational expense (opex) is much more attractive and less risky because of the tax implications. Opex like SaaS LIMS are fully tax-deductible in the year they are made, unlike capex, which are most often depreciated over five to 10 years.’

The GoInformatics platform includes LIMS, electronic laboratory notebook (ELN), project management, and resource management solutions that in combination act as a holistic platform for R&D and manufacturing in a range of sectors; these include animal health and nutrition, crop science, medical devices, pharmaceuticals, food and beverage, biofuels and contract research. The company’s aim was to develop an informatics infrastructure that was simple and easy to configure by SMEs, but also flexible enough to accommodate the more sophisticated requirements of larger institutions. ‘However complex those requirements, with SaaS solutions such as ours, clients get a dedicated support team that can address issues quickly, develop new functionality, or tweak the configuration, often within hours. When utilising a SaaS LIMS, the client will no longer need to submit support ticket requests into their often overburdened in-house IT team. More importantly, they won’t have to wait days or weeks for issues to be addressed,’ Medina pointed out.

Experimenting in the cloud

The SME sector has been quick to embrace the cloud culture, and the big industry players, while slower to transition, are now also implementing cloud solutions at a faster pace, Medina continued. ‘It’s been an educational effort to get the multinationals to move into the cloud, but they have experimented by adopting SaaS for non-mission-critical activities. Success in these areas has encouraged the adoption of cloud and SaaS solutions for more critical functions, such as LIMS or ELN, which handle sensitive data.’

Not for every business

Core Informatics offers a complete package of cloud-based LIMS, ELN, SDMS, and Collaboration modules to support biologics workflows and small-molecule drug development, but customers also have the option of deploying the Core Informatics software at their own facilities, Uzzo explained. ‘It makes sense for us to offer this option. Not every business wants to make a move into the cloud, perhaps because a cloud platform would need to co-exist and integrate with a complex array of other applications that are running on-premises. In some instances it’s easier to facilitate that integration by deploying our solution on our client’s servers, rather than try to marry one cloud solution with a raft of on-premises systems.’

Addressing potential deployment issues, Autoscribe’s Matrix Gemini LIMS is built on a dual web and Windows interface, and so can be run on a conventional desktop or via a web browser. This gives the firm the flexibility to address just about any client requirement with respect to how and where it is deployed, Boother comments. ‘Our Matrix Gemini LIMS can be installed inside the client’s firewall within their own informatics infrastructure, or hosted by a third party server specialist. We hosted our first system about six or seven years ago, and that partnership, based on a quarterly licence fee, has just been renewed again.’

Hosted services for sensitive data

In fact, most of Autoscribe’s clients have, to date, shown relatively little interest in having their LIMS deployed from a cloud server, or financed as SaaS, Boother noted. Scientists do still perceive cloud security as an issue, he believes, and this is one reason why, from Autoscribe’s perspective, the uptake of hosted services for informatics platforms that handle sensitive data has been relatively slow. ‘It is only more recently that some of our US-based clients have started to take up the option of hosted software as an alternative to on-premises implementation. Customers in the UK are following on slightly behind those in the US.’

Autoscribe uses tried and trusted third-party server specialists in the US and UK to serve clients who do opt for hosted platforms. ‘Offering Matrix Gemini LIMS as both a web and desktop solution means that cloud isn’t something that we see as make or break when it comes to attracting new clients,’ Boother stressed. ‘It’s effectively the same product however it is deployed, whether hosted on the client’s own server or through a third party server. What is far more important, from our perspective, is the functionality of our LIMS, its architecture, and how we can meet all the client’s informatics requirements, across its organisation. We just wait to be instructed on how to implement it.’