Aleksandrina Goeva returns from Toronto to Sofia to search for new medicines with AI
After building her own research team at the University of Toronto, Dr. Aleksandrina Goeva is coming back to Bulgaria. At INSAIT, Sofia University, she will lead a new research direction. Her team will use AI to predict how diseased cells respond to different compounds and which of them are worth testing as future medicines.
The possible combinations are so many that even a well equipped lab can test only a small share of them. Where should scientists start? This is where computer models can help. If they predict which compounds, alone or in combination, would help cells return to normal function, researchers can check the most promising ones first. The team will also study whether changes in cells can signal earlier that a disease is developing.
For Goeva this is also a return to the university where she studied. She graduated from the Sofia High School of Mathematics and then studied applied mathematics at Sofia University. At Boston University she earned a master's in mathematics and a PhD in statistics. She spent seven years at the Broad Institute of MIT and Harvard. Later, in Toronto, she built a team in machine learning and computational biology.
Among her work is the HiDDEN method, described in Nature Communications in 2024. It lets scientists tell cells affected by disease apart from healthy ones in the same sample. With it they noticed cell changes that had been missed in studies of blood cancer and of damage to the protective barrier between the blood and the brain. Goeva has also published in Science, Cell and Nature Biotechnology. According to Google Scholar, her work has been cited more than 7,200 times.
INSAIT announced the new research direction on November 9, 2026. The institute has not said what the budget will be, how many people will work on the team or when results are expected. Between choosing a promising compound and approving it as a medicine lie laboratory studies and clinical trials.
Sources
Told by Aurora, a language model. Every story is reviewed by a person before publication.


