Vacancies

The exciting research and development topics of the chair are looking for motivated postdoctoral fellows and students (PhD, MSc, BSc) who can participate in related projects.

The main research interests include the following areas:

  • artificial intelligence and machine learning in chemistry
  • cheminformatics
  • drug substance design and characterization
  • materials datascience
  • computational toxicology and chemical safety
  • quantum and computational chemistry

You can further explore the main directions on the private page and also from already published research papers.

Send expressions of interest together with an application and CV to Prof. Dr. Uko Maran (uko.maran@ut.ee).

It is currently possible to express concrete interest in two junior researcher-doctoral fellow positions and topics, both of which have an indicative start date of September 1, 2026 (flexibility is possible) and the application deadline is in May-June 2026.

Position 1

Identifying Chemical Hazards with Artificial Intelligence Methods

Traditionally, chemical risk assessment has relied on animal testing. However, growing ethical and economic considerations have driven a shift toward New Approach Methodologies (NAMs), which enable toxicity prediction using in vitro and computational approaches. Artificial Intelligence (AI) is foreseen as a new possibility for implementing computational NAMs, which are emerging as a set of tools for estimating chemical properties. Therefore, the primary focus of this doctoral study is the feasibility of AI-based methods for hazard identification of endocrine-disrupting chemicals. The study will utilise large and chemically diverse datasets to model androgen receptor binding. The research will cover alternative AI-based representations of molecular descriptors, particularly SMILES embeddings derived from specialised language models and deep neural networks, and transfer learning for quantitative structure-activity relationships (QSAR) to enhance model performance. In addition, the models developed within the present doctoral project will be published in accordance with the FAIR principles to ensure they are reusable for the scientific community and regulatory users. This work also aims to bridge the gap between AI tools and chemical risk assessment, advancing efficient and sustainable chemical safety science.

Position 2

Artificial intelligence approach for analysing and modelling the properties of ionic liquids

Ionic liquids have gained strong interest in the chemical industry and research to replace traditional volatile organic solvents as safer or better alternatives in synthesis and catalysis, electro chemistry, separation and extraction chemistry, analytical chemistry, biological and materials engineering, including electrospinning of polymer mats. The proposed doctoral research is aimed at the effective modelling of ionic liquids in very diverse datasets with a focus on multicomponent systems. Particularly, it is dedicated to the modelling of ionic liquids, to aid in understanding physicochemical interaction mechanisms when chemicals enter an ionic liquid and to model and predict such interactions. For this, the in silico approaches, artificial intelligence and machine learning will be used. Such methods allow bridging the lack of knowledge about chemicals when little or no experimental data are available. The qualitative and quantitative structure-property relationships (QSPR) approach faces new challenges due to the increasing diversity of chemical compounds measured and the need to predict properties of new chemicals for decision support. The methodology also allows fine-tuning the properties of ionic liquids for specific applications by introducing small changes to the chemical structure of the components. In summary, this doctoral thesis will investigate the complex relationships between ionic liquids’ properties and their chemical structures, eventually leading to the development of novel approaches for designing new ionic liquids.