The Hyve is a 30 person open source bioinformatics services company from Utrecht, Netherlands, and Cambridge, MA, USA. Not so long ago, we invited them to write a blog post on Open Source Technologies for Precision Medicine. Today, DNAdigest interviews Kees van Bochove, the CEO of The Hyve.
1. Could you give us a short introduction on The Hyve?
The Hyve is a small but fast growing global company that is specialised in professional support for open source software for bioinformatics, clinical and translational research. The company headquarters is in Utrecht, The Netherlands and we have a satellite office in Boston, USA, at Cambridge Innovation Center. The team (currently about 30 people) consists of a mix of bioinformaticians, software developers, clinicians, statisticians and project managers.
2. What is your background, interests and the role in the organisation?
My education is in computer science and bioinformatics, but I’ve always had a broad interest. Next to my computer science education I also graduated in primary school education, and I occasionally perform as a classical music singer: I recorded a CD with Schubert’s Lieder and I sometimes sing Bach cantata solos. I have taken an interest in biology and later medicine and I am engaged in patient organizations.I am the founder and CEO of The Hyve. My initial drive came from working with open source software in an academic context, where I saw that software is often abandoned after the paper is published or the consortium ends. I thought that there is a possibility to take advantage of these academic innovations in a corporate context, which is why I founded The Hyve to bridge the gap by providing professional services for open source software.
3. What are the mission and aims of the company?
The mission of The Hyve is to support pre-competitive collaboration in life sciences R&D, by leveraging open source software. We try to achieve that by applying our core values to our software and services: sharing, re-use and specialization. One of the long term goals of the company is to one day have a complete and robust open source software stack for the complete translational medicine field, from data capture at clinical sites, via bioinformatics methods and by wearables, via research insights generated by systems biology and medicine all the way to the application of data and knowledge in the clinic and at home.
4. What are the main activities and services of the company?
The Hyve offers professional services such as requirements analysis and project definition, data loading, software development and application support. We support several open source platforms for translational medicine, including tranSMART, i2b2, cBioPortal, OpenClinica, XNAT, Galaxy and Arvados, but also often go back to the basics and make use of more general data science and cloud provisioning tools such as R, Python, Jupyter, OpenStack, Docker, Mesos and Apache Spark. Our real strength is the combination of domain knowledge in bioinformatics and medicine with an advanced understanding of the modern data science tools and platforms and a solid project management approach.
5. The Hyve supports the open-source movement. Could you elaborate on this?
Open source is a legal concept: it is a way to allow other persons and companies to use the software or data that you create for free. There are various open source licenses; some have specific conditions such as in return demanding that you share any improvements you make to software under the same license.Open source has always been a mode of working in the bioinformatics community, for example in the early days of EMBL/EBI, but lately has seen a lot of interest in the pharmaceutical sector as well, as a means to have a concrete way of pre-competitive collaboration. It is also a good way of preserving investments done via e.g. European subsidies such as Horizon 2020.Unfortunately the exact legal terms of open source are often poorly understood, for example the pervasive but false notion that the GPL license is ‘viral’. Another critical,but often poorly understood aspect of open source is that the community behind it is much more important than the software itself.
6. What movements or organisations do you collaborate with?
We collaborate with several other organizations that also have a mission around pre-competitive collaboration in this sector, such as the Pistoia Alliance, its spinoff tranSMART Foundation and GA4GH. We also collaborate a lot with academic institutions, such as the Imperial College Data Science Institute, the Harvard Medical School department of Biomedical Informatics and the Netherlands Cancer Institute. Finally, we participate in consortia such as CTMM TraIT in The Netherlands and various Innovative Medicines Initiative consortia.
7. What is your vision of the problem of limited data sharing (in research in general and in genetics/genomics in particular)? How should it be addressed?
For research data, only the funders of this research, such as the European Union and various national initiatives, can really enforce sharing. For example, the NIH has mandatory repositories such as dbGaP to which NIH funded research data has to be submitted. Many funding organizations realise this and are now putting policies in place to enforce this.Data sharing in the genomics area is, as Ewan Birney put it at the last Global Alliance for Genomics and Health meeting, of crucial importance to advance medicine. However, there are real barriers and limitations when it comes to sharing medical data, especially in Europe. Some countries, such as Germany, have very strict privacy laws, which we encounter in our consultancy projects. I’m personally engaged in a number of initiatives on a global or European scale (e.g. IMI EMIF and GA4GH) as well as in The Netherlands (e.g. the Dutch Techcentre for Life Sciences and StartupDelta, with NeelieKroes) that will hopefully in the long run build a global infrastructure for responsible sharing of health data. I’m convinced that we can only deal with this by joining forces across many organizations.