Doctoral Candidate Position 9 - Bayer A.G. Germany
AI-driven data analysis, target identification and treatment development in PD
Ilektra Mitsi holds a background in Electrical and Computer Engineering and a Master's degree in Biomedical Engineering from Aristotle University of Thessaloniki, where she specialized in precision medicine, computational neuroscience, and machine learning applications in healthcare. At Pfizer, she led cutting-edge AI research initiatives including developing protein language model workflows integrating transformer architectures with AlphaFold2 to predict how viral mutations affect drug resistance, creating digital patient twin models for oncology drug development and repurposing, and spearheading bioinformatics pipeline automation using Nextflow and Airflow for RNA sequencing and CRISPR screening analyses. Her pioneering work combining multi-omics data (RNA, protein, and variant information) to predict molecular consequences of genetic mutations directly shaped her vision for integrated AI systems in drug discovery. Currently, Ilektra is a PhD student within the MSCA-funded BICEPS network, where her research builds on this foundation to develop an AI co-scientist platform that bridges computational prediction with biological understanding in neurodegenerative disease research.
Description of project
Parkinson's disease (PD) is a progressive neurodegenerative disorder characterized by the loss of dopaminergic neurons in the substantia nigra, leading to motor symptoms such as tremors, rigidity, and bradykinesia. Recent research has highlighted the complex interplay between the nervous and immune systems in PD pathogenesis, emphasizing the need for integrated approaches to study neuro-immune interactions. This project aims to research best practices for, and develop novel collaborative no-code analytical platform specifically designed to explore and analyze neuro-immune interfaces in the context of PD and other neurodegenerative diseases.
The research will first focus on full integration and making easily accessible evidence from both user-provided data and publicly available biomedical databases and literature sources, adhering to FAIR (Findable, Accessible, Interoperable, and Reusable) principles. This will be followed by the representation of data in a knowledge graph, enabling sophisticated reasoning and hypothesis generation. Finally, the project will establish explainable AI-powered analytics, reasoning, hypothesis generation, and ranking pipelines about targets, pathways, and potential treatments.
The platform will incorporate various data types, including genetic, biochemical, and clinical information, to provide a comprehensive view of the neuro-immune landscape in PD. By leveraging advanced machine learning techniques and natural language processing, the system will be able to analyze and integrate results from multiple experimental studies, literature sources, and other relevant data repositories. This multidisciplinary approach combines techniques from computer science, bioinformatics, and neuroscience to create a powerful tool for PD research.
Expected outcomes of this project include determining the precise best practices for multi-layered analysis of targets involved in neurodegenerative diseases and developing a system that can help scientists collaborate, review, analyze, and integrate results from diverse sources.
Ultimately, this work could lead to the identification of novel therapeutic targets and potential treatments for PD by facilitating a deeper understanding of the complex neuro-immune interactions underlying the disease. The collaborative nature of the platform will foster interdisciplinary research and accelerate the pace of discovery in the field of neurodegenerative diseases.