The Institute of Molecular Medicine Finland was established in 2006 and forms a part of EMBL and the University of Helsinki. The department is heavily focused on translational research. The institute is a melting pot where scientists from a range of disciplines come together to tackle grand challenges in precision medicine.
Aino Palva, NGS Service Expert and Laboratory Coordinator at the University of Helsinki introduced the iCAN project. This project was an enormous undertaking that capitalised on Finland's opportunities for big data by combining molecular profiling with digital health data to bring it closer to the patient. Palva honed in on the whole exome sequencing and RNA sequencing elements of the project. She added that the goal was to transition from semi-automated to fully automated lab processes to cope with the high throughput required by the iCAN project.
Now let’s look at the numbers, Palva explained that the plan was to sequence up to 15,000 cases until 2026 which roughly translates to 96 – 384 DNA and RNA samples per week, equating to 6 -12 terabytes of data per week. Therefore, it is clear that automation is essential for this project.
The former system relied on prepping the 96 samples manually or with the Hamilton system. However, the bead purification steps were automated on the Biomek system. Palva noted that the manual processing was painstakingly long and highly inefficient. The iCAN project calls for 96 + 96 samples and the goal was to prep and sequence the samples within one week.
The workflow began with a DNA and RNA extraction unit, HiPREP. Here, the team extracted and normalised the RNA and DNA from Finland’s biobank. Then Illumina’s DRAGEN platform performs tumour analysis on patient tumour DNA. For RNA prep the same platform is used to perform sequencing and analysis. The results showed that it only takes 6 – 20 minutes to process each exome and 15 – 30 minutes per sample to process. This is much more efficient than the former process.
As always, there were several challenges that cropped up. From Palva’s perspective, ensuring backup instrumentation was the main concern but she explained that by optimizing automation setups the system worked well. Her team also had to manage sequencing capacity and data storage. The lab had to address these issues to maintain efficiency and meet project goals.
To conclude, automation significantly increased throughput, improved on-target ratios, and uniformity in target capture in the iCAN project. Furthermore, it freed up scientists’ time for data processing.
