The project is dedicated to the advancement of computational methods for automatic language identification. Speech user interfaces are becoming more popular in everyday life. However, almost all the current commercial products using speech interfaces have to be manually set to understand a certain language as no automated language identification methods are used. The project is situated at the department of Digital Humanities of the University of Helsinki and the industrial partner is Lingsoft Oy. Lingsoft develops and uses their own speech recognition solution for a number of different use cases, the most important from a current business perspective are subtitling for television and medical dictation.

Cancer immunotherapies have been life-changing or even curing for some patients, however it’s still unclear why only some patients respond to immunotherapies. In this collaboration project, scientists from the University of Helsinki, Helsinki University Hospital, and Finnadvance are trying to solve the problem surrounding immunotherapy predictability by creating a microfluidic tumor-on-a-chip and modelling the human antitumor immunity. The platform combines patient’s own cancer and immune cells and could help to functionally predict the patients immunotherapy response in the future. A key element in this study is to increase the patient population benefiting from current and future cancer immunotherapies, as well as to increase the cost-efficiency of cancer care in Finland and beyond.

The University of Helsinki and Orion Pharma teams combine state-of-the-art flow chemistry technologies with machine learning for efficient synthesis of small-molecular compounds for the drug discovery projects. Our team is the first in Finland to combine synthetic organic and medicinal chemistry as well as catalysis with computer science. This benefits greatly the industrial host in the process optimization and identification the most facile synthetic routes of the target compounds. The post-doctoral researcher carries out ambitious and rewarding research and is supervised by a team of five leading industrial and academic scientists. The project pioneers also as an initiative on how to train the next generation of chemists, conversant in modern chemical synthesis and computer science.

Climate change is accelerating and greenhouse gas emissions are reaching an alarming level. Forests play a crucial role in modulating the climate, though some evidence reported the warming effects of forests. The effect of forests on local climate can be determined through the interaction among multiple mechanisms such as photosynthetic uptake of carbon, land surface albedo, and evapotranspiration, either of which may cause cooling or warming effects. As such, the climate impacts of forest restoration will vary, based on regional differences. These knowledge gaps suggest an urgent necessity for a better understanding of mechanisms that determines the impacts of forests on the local air temperature to guide effective forest restoration decision making at regional and global scales.

In this project we will build an electrospray based thermal desorption chemical ionization mass spectrometer, and use it to measure the chemical composition of sub-100 nm aerosol particles. We will build a CFD model of the ionization inlet, and based on that build a new and better inlet. After that we will study the suitability of new ionizations chemistries for detection of atmospherically relevant molecules. We will calibrate the new inlet and use it to measure the chemical composition of atmospherically relevant sub-100 nm particles.