In 2015 Dr Marinka Žitnik gained a doctoral degree at the Faculty of Computer and Information Science (FRI), receiving for her work the gold Jožef Stefan emblem, which is awarded each year to the most outstanding doctorate by the Jožef Stefan Institute (IJS). She is currently employed as a post-doctoral researcher at Stanford University, the eminent institution in the USA, and at the Bioinformatics Laboratory at FRI, where she is studying the approaches of artificial intelligence and machine learning to detect knowledge in large biomedical data sets.

We talked to her about her work abroad, her doctoral studies and research work.  

1. Two years ago you successfully completed doctoral studies at the Faculty of Computer and Information Science in Ljubljana. What was it that made you decide to pursue doctoral studies? What did you have to consider, and what was decisive?

I have always been fascinated by computer science and mathematics. This fascination, together with my research experience gained during my undergraduate course led me to further study. I am also a very curious person anyway, and I am always trying to learn something new about the world around me. A scientific method and systematic approach to solving challenges seem to me the driving forces of technological progress. Equally, I believe that in-depth knowledge of computer science enables us to apply critical judgement to complex systems and to be able to ask the right questions at decisive moments. This personal view has guided me through my entire education and played a key part in me deciding to pursue doctoral studies.

2. The path to a doctoral degree can sometimes be extremely long, but you completed yours with above-average speed and at the same time published numerous articles. What can you tell us about your experience as a doctoral student?

My experience was great! I had a lot of freedom in my research, and I was able to get involved in areas that inspire me. Furthermore I had an outstanding mentor, superb research colleagues and abundant opportunities to garner experiences at distinguished institutions abroad. I have always tried to do a lot of work and to do it to the best of my ability. So this explains how it took me less than three years from enrolment to submit my doctoral thesis.

3. Even during your doctoral studies you were a guest at several distinguished foreign institutions, and now you are a post-doctoral computer science researcher at Stanford University in the USA. What are the differences between studying or research in Slovenia and abroad?

Being a guest at a foreign university has been very important, often this catalysed new ideas, it started or enhanced collaboration on research and offered an open door for future opportunities. I am grateful to the professors who received me into their groups so agreeably at the University of Toronto, Imperial College London, Baylor College of Medicine and Stanford University. In Slovenia and abroad I encounter top researchers who are pre-eminent in their fields in the world. Yet it seems sometimes that society at large in Slovenia is not as favourably inclined towards the research environment as elsewhere abroad, but perhaps this is just my subjective viewpoint.

4. In your doctoral work you were involved in the development of algorithms of artificial intelligence and machine learning that enable conclusions to be drawn in large and multifaceted data systems. Tell us something about this.

Modern technological, scientific and biological systems enable the capture of almost unlimited quantities of heterogeneous data. These data describe systems on multiple strata and from different viewpoints. But the data are often set out in entirely different data domains, which presents challenges for machine learning and artificial intelligence algorithms. My doctoral thesis is based on the premise that heterogeneous data can be “organised”, so that we can learn appropriate mapping among data dimensions. The bottleneck that separates us from a better understanding of data is in fact the recognition of information that establishes a link between data that are apparently unconnectable. This information serves as a kind of “adhesive matrix” for designing a mosaic from all the tiny data tiles.

I proposed several promising and powerful algorithms for learning through data fusion. These approaches attain a high predictive accuracy and offer a general solution for constructing data mosaics from what are in principle unlimited numbers of data tiles. My colleagues and I used these approaches successfully in molecular and systems biology for predicting genetic functions, ranking promising genes for further biological research (for instance, we predicted new functions of nine genes that were confirmed by experiments in the biological laboratory), discovering patterns of connections between diseases, identifying the toxicity of medications and analysing mortality.

5. Currently you are a post-doctoral researcher at a distinguished foreign university, where you are continuing your successful career. What are your plans for the future?

In the future, too, I want to continue my path in research and innovation environments that are excellent, dynamic, developmentally active and stimulating.

Enrol in a doctoral study programme. The deadline for applications is 2 June 2017. Apply online via the eVš higher education portal.