In the Spotlight - Meet Maryna Chepeleva
At Women in AI we’d like to inspire others by featuring women role models that are making a difference in AI. We aim to empower women who are working in AI by highlighting their success stories. This way we hope to inspire other women and girls to get into STEM-related fields. In our series ‘In the Spotlight’ we shine a light on an expert in the field and today we’d like you to meet Maryna Chepeleva.
Author: Sharadhi Alape Suryanarayana
About Maryna
Maryna is a PhD student in Bioinformatics at the Luxembourg Institute of Health, where she is a part of the Multiomics Data Science Research Group. Her expertise is multi-disciplinary, in data analysis, machine learning, and bioinformatics, given her strong academic background and industry experience. Consequently, her research focuses on developing innovative computational tools for addressing complex biological challenges in cancer research. By integrating several data modalities using data analysis and AI, her work aims to enhance the understanding and treatment of cancer diseases.
The diverse facets of data analytics in cancer research
Maryna’s primary focus is on developing unsupervised methods for extracting meaningful signals from complex multi-omics data. Multi-omics data includes the datasets generated by analyzing various “omics” layers in biological systems such as genomics, transcriptomics, metabolomics, etc. Her aim is to identify the latent structures within the data that are not apparent through traditional analysis methods. These extracted features can serve as powerful predictors in supervised models, enabling more accurate patient classification and predictive analytics.
In her key project, she is working on the integration of metabolomics and transcriptomics data to uncover biomarkers and molecular mechanisms underlying drug resistance in melanoma. Using unsupervised learning techniques, such as consensus Independent Component Analysis (consICA), she is able to decompose complex biological data into independent components. These components reveal hidden patterns and associations that are critical for understanding disease pathology and developing personalized treatment strategies.
Her work also includes the development of the consICA R/Bioconductor package, which provides a framework for robust reference-free deconvolution of heterogeneous data. This tool enables researchers to identify differentiated biological programs and stratify the samples based on molecular profiles without the need for prior knowledge. Insights gained from these analyses are not only useful for classification purposes but also for explanatory analyses that can drive the discovery of novel therapeutic targets.
Furthermore, previously she contributed to the creation of the ORFhunteR package, which automates the identification and annotation of open reading frames in human mRNA sequences through pre-trained classifiers. This tool supports genomic research by streamlining the analysis of genetic sequences, thus facilitating the discovery of new genes and the understanding of their functions.
Maryna’s ambitions: From understanding of multi-omics data to advancing AI-driven personalized medicine
Integration of AI with medical research comes with multiple opportunities to contribute to a better world. From understanding disease mechanisms to developing personalized medicine and reducing healthcare costs, Maryna has her journey mapped from highly specialised research goals to making global impact. Currently, her focus is on her research where she is developing AI-based methods to integrate and interpret multi-omics data which enables the identification of key molecular features. This in turn helps in understanding disease mechanisms and improving patient outcomes. She is also keen on developing R packages that provide new AI-based methodologies that extend possibilities of data analysis for the Bioconductor community.
Looking at the bigger picture, Maryna feels that her research on developing methods for the understanding and integration of multi-omics data contributes to a growing body of knowledge that supports the creation of more advanced models and tools. Moving ahead, she wants to further the spread of AI-driven personalized medicine approaches and thus, improving patient outcomes and reducing healthcare costs.
Being Part of the WAI Awards
Curious to learn more about Maryna, her work and the amazing work our other Women in AI Award nominees have done? Join us on February 7th at the Women in AI Gala, where the official Award winners will be presented. Come and celebrate all these remarkable women forging the path on making a significant impact in AI across the Benelux region. We can’t wait to see you there!