UPWr Doctoral School: prof. Szyda on bioinformatic analysis

Prof. Joanna Szyda from the Department of Genetics at the Faculty of Biology and Animal Breeding combines genetics, statistical analysis and mathematics in her research. As she explains, today's genetics are no longer the legendary peas of Mr. Mendel, which schools use to teach about how traits are passed on.

- Our knowledge is reaching deeper and deeper, but this means that we also need tools that will allow us to group, analyze and draw conclusions from collected data. These tools are mathematical statistics and computer science, because it is difficult to imagine modern genetic research, which is increasing its precision, without software that will allow the obtained data to be analyzed. It cannot be done "by hand", on paper, because the current biological data is often large-scale data that fits into the fashionable category of big data or data mining - explains Prof. Joanna Szyda.

As emphasized by the Vice-President of the Animal Science and Fisheries Discipline Council, the Doctoral School at the Wrocław University of Environmental and Life Sciences is a place where young scientists looking for a challenge can find their way into great research teams.

– Our doctoral school is open to ambitious people. But it also offers many possibilities - underlines prof. Szyda.

One of the more ambitious students is Krzysztof Kotlarz, the winner of the first edition of the "Preludium Bis" National Science Center grant. The PhD student of the UPWr Doctoral School is also a first-year master’s degree student, as such opportunities are provided by the "Preludium Bis" grant. His doctoral thesis is based on the use of artificial intelligence to analyze large-scale data – that is, big data – in context of genetic analysis of cattle population. The first part of the project involves the use of deep learning to create a classifier of correctly and incorrectly identified polymorphisms based on data of whole DNA genome sequences, the so-called DNAseq. The second part of the analysis concerns the classification of cows as susceptible and resistant to mastitis, based on genotypes of polymorphisms also obtained from whole genome sequences. The last stage of his work is to create a cattle breed classifier – also based on data from DNA genome sequences.

– Krzysztof also works on analysing other large-scale data. This is an additional project. We use a classic biostatistics tool – linear mixed models – to calculate the effects of genes that have experienced expression changes under heat stress. This project uses another type of large-scale biological data - expression of the whole exome, the so-called RNAseq of rats as a model species - explains Prof. Joanna Szyda.

Currently, the doctoral school, in which the Vice-President of the Animal Science and Fisheries Discipline Council is one of the supervisors, has submitted a doctoral project titled "Analysis of the soil’s microbiome in industrially polluted areas", i.e. a comparison of species and quantitative composition of the soil’s microbiome.

- The analysis will cover soils from post-industrial areas, for example, from copper smelters in Legnica and Głogów, and from corresponding morphological and type of use attributes in soils from areas not covered by industrial activity of smelters, serving as a control group - says prof. Joanna Szyda, explaining that depending on the financial possibilities of the project, the plan is to obtain region sequences of the gene which encodes the 16S ribosomal subunit or the sequence of entire genomes of soil microorganisms, using shot-gun technology.

Work on this project, in the field of biological sciences, includes bioinformatic analysis of DNA representing the microbiome of soil samples and statistical analysis of data.

In the field of agricultural sciences, however, we are waiting for a candidate with whom we will be able to develop the latest statistical methodology for predicting the genetic value of bulls. For this purpose, we will use a so-called one-stage model, with all available sources of information, such as the performance of individuals and their genotypes.

- As you can see, modern agricultural sciences also involve knowledge of mathematical statistics, the ability to work with various open-source software and its installations, or to code your own programs in Python and R. The UPWr Doctoral School gives the opportunity to develop them, as well as to work in international research teams - emphasizes prof. Joanna Szyda, adding that thanks to the work of geneticists and the tools used by IT, animals reveal more and more secrets to scientists. This is also possible due to the doctoral theses carried out at Wrocław University of Environmental and Life Sciences.

kbk