2019-04-11

Fredrik Dyrkell, CTO, explains the power of the 1928 platform

Fredrik Dyrkell is the CTO of 1928 Diagnostics. His expertise is within software development, bioinformatics, user experience as well as quality and regulatory affairs. His daily job is to lead the software development team, and he feels that working at 1928 Diagnostics is a unique opportunity to use his knowledge to create a service that can make a real difference.

The company was founded to stop the spread of antibiotic resistance globally and to save the power of antibiotics. What we offer is a cloud service that analyzes resistance markers in bacteria from whole genome sequencing. We look for resistance markers within the bacteria and perform a relational analysis between samples. The relational analysis can be used to look for possible outbreaks at an early stage and also improve an active infection control.

With more data we can do more advanced analysis

Fredrik explains that the price of whole genome sequencing (WGS) has decreased dramatically the past years and it continues to do so. He believes that this will drive the adoption of WGS in clinical routine, and with more data the 1928 platform can do even more advanced analysis and learn more about resistance and pathogens.

And it is quite easy for hospitals to try out the platform for analysis, once they have started using WGS on bacterial samples. The output from the WGS is a set of files with the full DNA of the bacteria, and out of that data 1928 analyzes and extracts useful information.

Upload sample files and get the results in 15 minutes

A bioinformatician will start by uploading the sample files to the cloud-based platform and the analysis begins automatically. The results will be available within an average of 15 minutes. But what actually happens during the 15 minutes?

When the samples are uploaded there must be a quality assessment to make sure that we have sufficient sequence depth and the right information to do a correct analysis. The first part of the analysis is resistance prediction, that looks for genes and mutations correlating with the resistance. Then follows infection control analysis, where we create a phylogenetic tree with the relations between the samples, which can be used to detect outbreaks. The types of information you find in the results are for instance sequence types, virulence factors, and for the resistance markers we link to the research literature with a clinical relevance. All of this serves to improve the hospitals' infection control and proactivity in detecting outbreaks and hospital acquired infections.

Testing and verification for quality control and patient data security

Fredrik points out that the team puts a lot of emphasis on quality control, and that the team members are heavily invested in automated tests and verification that the platform meets the right criteria. The platform is also GDPR compliant, so hospitals can send patient data for analysis through the platform in a safe and compliant way.

For the future, the software development team is right now working on long-read sequencing and analysis of those, with hybrid technologies between long-read and short-read. We are also working on machine-learning models to predict antibiotic resistance using learning algorithms, as well as meta-genomic analysis where we are looking at typing bacteria within meta-genomic samples.

No upfront investment with a SaaS platform

In addition to the described benefits of using our technology, 1928 is a smart choice from a business perspective. Since the 1928 platform is a cloud-based software, it can operate at scale and with a high level of service. Using a Software-as-a-Service (SaaS) model there is no upfront investment for the hospitals, and you only pay for what you use.

Several hospitals in Europe and the USA are already now getting value out of the 1928 platform. So if you are starting to do WGS, we will be happy to provide the analysis for you.