Embedded Systems; Genomics; FPGA; Low Power; In the field analysis
Casale Brunet Simone, Schuepbach Thierry, Guex Nicolas, Iseli Christian, Bridge Alan, Kuznetsov Dmitry, Sigrist Christian, Lemercier Lemercier, Xenarios Ioannis, Bezati Endri, Towards in the field fast pathogens detection using FPGAs, in 2018 28th International Conference on Field Programmable Logic and Applications (FPL)
The rapid detection and identification of infectious pathogens are a critical need for healthcare in both developed and developing countries. Remarkable progress has been made in using genomic information to prevent and manage viral outbreaks. However, more than one billion people in resource-limited developing countries still suffer from one or more neglected infectious diseases, most of which are found in Africa and Southeast Asian countries. Infectious disease diagnostics and epidemiological surveillance systems in many of these countries are not as advanced as in developed economies, where continuous power, strong communication infrastructure, laboratory infrastructure, trained personnel and conventional sequencing technologies are available. This disparity has many consequences, both ethical and social. Ensuring advances in genomics applied to the health improvement of peoples’ living conditions in developing countries is an important contemporary challenge. The purpose of this project is to give access to worldwide health organizations and local administrations to a real-time embedded and battery powered surveillance solution, not bigger than a smartphone. This solution permits a rapid detection and identification of many diverse infectious disease pathogens by using embedded genomic analysis. To deploy this solution an embedded DNA sequencer should be used for retrieving the genome of each individual at the point of care. A potentially embedded sequencer is the MinION from Oxford Nanopore Technologies. Successively, each genome is analyzed in order to find the presence of known pathogen patterns using proprietary algorithms that have been developed by the Swiss Institute of Bioinformatics (SIB). This step is extremely computationally expensive and nowadays solutions require sending genomic data to remote analysis laboratories through cellular data networks which can be extremely slow and unreliable. The objective of this project is to revolutionize this part by providing an embedded and low-power solution that enables the analyses in-loco without requiring any expensive clinical equipment or other kinds of technologies. For this purpose, a multiprocessor-system-on-chip architecture that contains a reconfigurable logic (MPSoC with FPGA) will be used. This kind of chip provides low-power massive parallelism but with the cost of being difficult to program. The scientific work achieved during the PhD studies of the candidate has demonstrated that MPSoCs can be easily and efficiently programmed exploiting high-level dataflow programming paradigm and models. Finally, his current research work within the Swiss Institute of Bioinformatics (SIB) has already proved how genomic analysis algorithms can be effectively implemented on FPGAs platforms.