01 May 2019
Share this article

We want to be the Amazon for biosamples and data

Select the biosamples or patient data you are looking for and get them delivered to your doorstep, whenever and wherever you need them – just like Amazon or Bol.com. For many life-scientists, that used to be only a dream.

With BBMRI-NL’s ‘Catalogue’ biobank database and the ‘Podium’ request portal it’s now a reality, giving researchers the opportunity to search through hundreds of biobanks and databases in the Netherlands and request the biosamples and data they need. Catalogue has recently been improved via a more user-friendly interface and direct links to the Podium request portal.

Researchers spend a lot of time trying to find the biosamples and data they need. “In the past, you only got access to the samples at the end of your corridor. Nowadays, research is interdisciplinary and involves multiple institutes.” Prof. Morris Swertz is one of the driving forces behind Catalogue and is explaining why it was developed: “It’s the catalog-of-catalogs for biological and clinical biosamples and data. You can search through the stock of all associated catalogs for specific disease types, genome-wide association study data, types of biomaterials and many other search variables. It makes finding samples and data much easier, and if you find it’s already there you don’t have to start your own collection of samples or your data cohort.” Swertz is one of the driving forces behind Catalogue. While ‘Catalogue’ makes data findable, ‘Podium’ is the on-line tool for requesting and accessing the biosamples and data you need.

The devil is in the detail

The sample request process was first semi-automated for PALGA, the Dutch national network and registry of histo- and cytopathology. The PALGA registry contains data from many different laboratories, each of which has curated archives of clinical data and biological samples as part of their regular healthcare activities. Collectively this is valuable material for research, but it used to be difficult to obtain explains Prof. Folkert van Kemenade, a pathologist involved in the PALGA portal from the beginning. 

“The sample request process used to be long and was often done manually. You needed to find an organization that has the data you want, send an application form, wait for approval, and then wait for the data to be delivered,” he says. “It’s not just the waiting time. You needed to know where to find the type of data you need and follow the status of your applications, especially when multiple organizations or biobanks were involved.”

Left: Prof. Folkert van Kemenade

However, Van Kemenade also points out that developing a request portal to solve the problem is not just about building an ICT-tool. The organizational aspects are very complicated and need to be adequately addressed.

“Developing a portal involves many different steps, with different labs, local secretariats and different approval procedures and protocols involved.  You think you’ve considered every detail, but when you try to put theory into practice, you find the devil is in the detail,” he says. “But problems are there to be solved. The PALGA portal enables you to request pathology data and biosamples safely from all over the country. It was a big initial boost to the concept and development of Podium.”

Erik van Iperen, technical project leader for the Podium portal agrees. “The PALGA Portal truly inspired us. There was a clear need for a more generic portal for requesting samples from biobanks, images from neuroimaging centers, data from registries, and data cohorts,” says Van Iperen. “So, we started by contacting different user groups and translating their wishes into requirements for the Podium Portal developer, IT company the Hyve. Now, with Podium, organizations can easily manage the data-requests they receive, because requests are stored in one central system that supports track-and-trace. You don’t need to handle requests by telephone, email or fax anymore.”

A one-stop shop

Now that the new portal is up and running, Van Kemenade foresees more advantages for the Catalogue – Podium combination. “With Podium we’ve strived for uniform request procedures, allowing biobanks and data cohorts to learn from each other and researchers to make linked requests. For example, they can combine samples from different data sources at an individual level. Organizations then need to agree on the type of data, the process and the consent. And after that, the technical aspects have to be dealt with,” he says. “These collaborations result in an accumulation of experiences that researchers can draw on so that they don’t have to repeat the work.”

A Linked Request
Does an operation on the cervix lead to prematurity or effect other birth parameters? To answer that question, we wanted to link data on the health of the baby, stored in the Perined database, with data on the operation results for the baby’s mother, stored in PALGA. We managed to do this by using the services of Statistics Netherlands (CBS). It allowed us to couple records while preserving the privacy of the patients. The CBS was really helpful, and open to all kinds of research questions. 

The future of Catalogue

Catalogue does not contain privacy sensitive data so the information it contains cannot be linked to individuals. As a result, everyone can access it. It is also connected to other European catalogs so that researchers from other European countries can search it and vice versa.

Esther van Enckevort, technical project lead for Catalogue at the University Medical Centre Groningen (UMCG), emphasizes that Catalogue not only makes collections findable, but also ensures that all request procedures remain intact.

“In the NFU Data4LifeSciences project, which is included in Catalogue, we are encouraging UMCs to set up the digital research environment in such a way that metadata is included in the collections. For example, the type of consent that patients have given, because uncertainty about consent is a common problem that often precludes the reuse of data,” he says.

Left: Prof. Morris Swertz

According to Prof. Morris Swertz, Head of UMCG’s Genomics Coordination Center, the next step is to make Catalogue even better by generating more detailed descriptions of each sample, making it possible to locate more rare samples. “It will take Catalogue to the next level,” he says. “From a ‘telephone book’ to a ‘Yellow pages’ for biosamples, or better even, from the level of a dictionary to that of an encyclopedia.” In the meantime, he is actively looking to add more data collections to Catalogue – both existing collections at universities or hospitals and new ones.

Podium promises

Catalogue already contains a great many entries, with all its collections and organizations eventually being coupled to the Podium Portal. “Podium is now coupled with collections from Parelsnoer and BBMRI studies, and other registries, biobanks, cohorts and databases will soon follow. Before this summer, it will also be coupled to GO-NL, bios, and several metabolomics collections, with the PALGA portal following later,” says technical project leader Erik van Iperen. “The experiences of these users will help to further improve Podium, leading to better and more robust protocols. It really shows that Podium is set up by researchers for important research purposes.”

Erik is now busy approaching various owners of collections. “I am inviting them to make their data accessible, which they can do via the helpdesk of Health-RI,” he says. “And, of course, we encourage new organizations to sign up for both Catalogue and Podium at the same time.”

Erasmus Medical Centre’s Chair and Professor of Pathology, Folkert Van Kemenade, is enthusiastic about the future.  “I expect that by the end of 2020 most of the Dutch databanks will be coupled to Podium,” he says. “That will be all kinds of resources – the Nivel database for primary care data, surgical data from DICA, DNA sequences et cetera. It will then be possible to request image data, DNA-sequences, clinical data, and biosamples such as frozen tumor biopsies or feces, from different sources all through a single portal.”

Are you currently coordinating a sample or data collection and would like to add it to the BBMRI-NL Catalogue? If so, please contact the Health-RI  Service Desk. If you request inclusion in Catalogue, we will send you a ‘data integration manual’ to help you. Download the BBMRI-Catalogue form.

Facts and figures
* Catalogue and Podium have been developed within BBMRI-NL. 
* Work on Catalogue started in 2011.
* Catalogue currently (April 2019) covers 278 collections of all types and sizes. A few examples:
  • PALGA, which stores some 35 million samples
  • Lifelines, a population cohort with data from 165,000 participants
  • The Rotterdam Study, which comprises data from half a million healthy people

* On an average day there are around 10 searches within Catalogue.
* The beta-version of Podium was launched in December 2017.

People involved in building and promoting Catalogue and Podium

Esther van Enckevort (University Medical Centre Groningen): technical project lead for Catalogue and now coordinating its further development.

Erik van Iperen: technical project leader for Podium

Folkert van Kemenade (Erasmus Medical Centre): one of the initiators of the PALGA Portal and Podium.

Morris Swertz (University Medical Centre Groningen): principal investigator for Catalogue and an active user of the catalogue as a researcher.