The bioinformatics working group was founded in 2013 and was formalized as an official BeSHG working group (BelGenBioIT) in 2018. During these years, we have built a stable bioinformatics network within the belgian centers of genetics: sharing our expertise and discussing best practices and new developments, as well as specific issues linked to the application of bioinformatics in the context of accredited medical analyses. We would like to reflect on the future and to lay out our strategic view for the upcoming challenges in our field.
We first believe that our current network will be stronger if it fosters more collaborations internally and if it reaches to other bioinformatics experts working in diagnostics in Europe. Indeed, our analytical workflows need specific optimizations to scale to the demand of our medical laboratories. Moreover, our workflows will certainly improve if we continue sharing best practices and if we work towards the homogenization of our procedures and validation strategies. Working together on these challenges will contribute to improved steps in the diagnostic procedures and will indirectly lead to improved time to diagnosis and/or improved quality of diagnosis.
Secondly, we strongly support and encourage the use of open-source software in our analytical workflows. We believe transparency is an important component of the trust put in us by patients, and transparency and accountability are at the core of the open-source software model. Moreover, this strategy leads to an acceleration of the transfer of applications from research to diagnostics, which is beneficial to the patient.
Thirdly, there is a growing need for data sharing in the context of genomic data, and we advocate for a federated paradigm as a model for genomics datasharing. Classically, database models have been centralized but recent global efforts (GA4GH, Elixir) have worked towards a new model. Indeed, sensitive personal genomic data should follow a decentralized model where data remains in each center responsible for its analysis. There has been a growing effort in the development of decentralized data sharing software and we also believe this is the future of genomics data sharing.
Finally, one of the main challenges in the upcoming years remains that of reducing the data footprint of medical genomic analyses. Several solutions have been under development to reduce it by the use of compression algorithms. We have been testing and implementing these solutions in our facilities. However, a question that will need to be answered in a near future is the need to store this data. Given the GDPR and the development of new sequencing instruments, this stored data could be rendered useless or go against the will of patients.
In this context, there is an opportunity to work together towards achieving these goals with the objective of bringing WGS to the clinic. Several challenges will need to be met concerning (1) the homogenization of our workflows and validation strategies, (2) the optimization of current analytics software, (3) the sharing of pathogenic and non-pathogenic variants at a population level, and (4) the reduction of the data footprint. We think these challenges will be better tackled with a strong and collaborative bioinformatics network.
Member list :