The research of a group of bioinformaticians from Szeged has caught the eye of the Chan Zuckerberg Initiative. The Hungarian scientists excelled in the field of cell analytics aided by artificial intelligence.
Heti Világgazdaság reports that the Chan Zuckerberg Initiative is aiding a group of bioinformaticians from Szeged, led by Péter Horváth. The group has developed a method in cell analytics that is unique in the world.
The aim is to perfect this method to the extent where cells can be analysed to such detailed that the smallest changes in a system can be detected. This way, scientists and medics can get a better understanding of cellular biology processes and pathological lesions.
The focus of the group are phenotypes, which are unique to every organism. Phenotypes carry information that is just as vitally important as our genetic codes: the changes in phenotypes are telling of a possible fault in an organism. To understand precisely these subtle and dynamic changes in phenotypes, hypersensitive microscopes and advanced processing systems are extremely helpful.
One of these softwares helpful in discovering new phenotypes among billions of cells was developed by Péter Horváth (Institute of Biochemistry, Biological Research Centre, Hungarian Academy of Sciences, Szeged) and his team of scientists.
The microscope system developed by these scientists can cut around the cells that are to be examined precisely and take it out of the sample.
The achievements of the Hungarian scientists have drawn international attention. Frequently cited German researcher, Matthias Mann (director of the Max Planck Institute of Biochemistry and professor at the University of Copenhagen) and one of the youngest leading scientists in the field of proteomics and cellular biology, Lundberg Emma (professor at the University of Stanford and at the KTH Royal Institute of Technology) both got interested in working with the team.
The aim of the research deeply embedded with in-depth visual proteomics that gained the support of the Chan Zuckerberg Initiative (set up by Mark Zuckerberg and his wife, Priscilla Chan) is
to create an intelligent image processing system which is able to recognise every cell in human tissue,
separate cells with different phenotypes, retrieve several kinds of cell types and analyse all proteins gained from the retrieved cells quantitatively.
How exactly did the Hungarian team get the attention of the Chan Zuckerberg Initiative? They have participated in a bioinformatics competition not long ago, competing with almost 4 thousand teams. The task was to create an image processing software for microscopes that was able to identify every cell in pictures that the algorithm was not introduced to.
The research team led by Péter Horváth developed a program that is able to create artificial images with an AI algorithm, using a different AI algorithm to help the previous one identify objects on the images.
Their software, nucleAIzer recognised every single cell on the images, achieving the highest scores.
Source: hvg.hu
