Machine Learning and MALARIA SEROLOGY
Stakeholders
Fundação para a Ciência e Tecnologia, Portugal (Funder)
MI2 Data Lab (Host)
ANTIMALARIAL Antibodies CAN BE USED TO DETECT WHO IS LIKELY TO BE RESISTANT TO CLINICAL MALARIA OR had a RECENT MALARIA INFECTION. WE ARE developing tools based on MACHINE LEARNING TO IDENTIFY KEY ANTIMALARIAL ANTIBODIES THAT COULD BE USED in MALARIA EPIDEMIOLOGY AND VACCINE DEVELOPMENT.
Background
The current understanding of antimalarial immunity is undermined by the lack of reproducibility across studies. This problem can be explained by the complex life cycle, constant evolution, and high genomic diversity of malaria parasites. Another factor is the adoption of quantitative solutions whose results lack generalization. To solve this problem, we aim at developing new analytical pipelines combining statistical and machine learning methods to analyze public high-throughput antibody data.
André Fonseca is conducting this project under the supervision of Dr. Nuno Sepúlveda (main supervisor) and Dr. Clara Cordeiro (co-supervisor). André is based in the University of Algarve and he is fully funded by the Fundação para a Ciência e Tecnologia, Portugal. He is currently on a scientific mission at the MI2 DataLab group led by Prof. Przemyslaw Biecek to learn machine learning techniques in greater depth.
Expected Impact
The pipelines under development are meant to be deployed as tools to develop antibody-based diagnoses of recent exposure to malaria parasites and of protection aganst clinical malaria.