Filip Stefaniak

personal homepage


me me me me

I'm the organic chemist by training with seven years med-chem and CADD experience in pharma companies. Currently, in academia, I focus on the translation of computational algorithms to new bioactive compounds. I'm the Machine Learning enthusiast.


Scientific projects

See also: Scientific publications

Prediction of compound activity and toxicity

I've created the pipeline for fast and reliable prediction of compound activity and toxicity. This approach was ranked in the top five most accurate predictions in the 2014 NIH Tox21 challenge. For more information, see: Prediction of Compounds Activity in Nuclear Receptor Signaling and Stress Pathway Assays Using Machine Learning Algorithms and Low-Dimensional Molecular Descriptors, Front. Environ. Sci., 01 December 2015 |


A new scoring function for predicting RNA ligand interactions

I develop the new scoring function for predicting RNA - small molecule ligand interactions. With help of deep learning methods.


Computer Aided Drug Design

I'm involved in several projects focusing on the developement of new bioactive molecules. Recently published paper: Discovery of TRAF-2 and NCK-interacting kinase (TNIK) inhibitors by ligand-based virtual screening methods, Med. Chem. Commun., 2015, 6, 1564–1572

binding pocket

Currently I'm involved in the developement of the new, selective inhibitors of flavivirus methyltransferases (Yellow Fever Virus, Dengue Virus, West Nile Virus and Zika Virus).

Non scientific projects

Runs statistics

A collection of statistics of various runs and road races (5, 10, 21, 42 km) in Poland (in polish).

violin plot

Professional skills

Drug developement

Computer Aided Drug Design

Data analysis and visualisation

Computer skills

badge badge badge

Running and Triathlon - Personal Bests