I-GAME: Integrated Genomics and AI as a tool for Malaria Elimination in Brazil

Malaria, caused by Plasmodium parasites, continues to be a major global health concern, with millions of cases and hundreds of thousands of deaths annually. In 2023, Brazil reported over 140,000 cases, marking it as the country with the highest malaria burden in South America. Notably, there are significant data gaps, particularly in high-risk regions such as indigenous communities and gold mining areas around the Amazon. Efforts to control malaria worldwide are further complicated by the emergence of Plasmodium drug resistance (DR), especially against artemisinin-based treatments. While resistance to artemisinin has primarily been observed in Southeast Asia, there is concern that similar issues may arise in other regions with comparable transmission dynamics, including parts of South America like the Brazilian Amazon. The generation and analysis of Plasmodium genomic data are critical for identifying DR mutations and understanding transmission patterns, including the cross-border movement of strains. Advanced genomic techniques, such as whole-genome sequencing (WGS) and targeted gene amplicon sequencing (AMP-SEQ), are used to identify species, DR mutations, and genetic diversity. Platforms like Oxford Nanopore and Illumina provide detailed clinical and epidemiological insights, thereby enhancing surveillance strategies. However, the effective utilisation of extensive genomic datasets is often hampered by a shortage of bioinformatics expertise and advanced informatics tools. Developing AI-driven informatics tools, such as the Malaria-Profiler software, is crucial for the rapid analysis and interpretation of WGS data. These tools can provide actionable insights into species identification, DR profiles, and geographic origins, which are essential for guiding clinical management, surveillance efforts, and public health interventions, particularly in data-limited regions like Brazil. Leveraging a well-established collaboration in malaria epidemiology, with extensive field site access and expertise in genomics and AI methodologies, the London School of Hygiene & Tropical Medicine (LSHTM) and the Institute of Biomedical Sciences at the University of São Paulo (ICB-USP) aim to further enhance these informatics tools. The project seeks to integrate AI models to continuously update mutation libraries and improve the predictive accuracy for species identification, DR profiling, and geographic profiling, alongside other genomic information that could support the National Malaria Control Programme (NMCP). This initiative includes conducting WGS/AMP-SEQ in Brazilian malaria hotspots to better understand genetic diversity and inform strategies for disease control and elimination in the country. The integration of AI methods with genomic data for parasite profiling has the potential to revolutionise malaria control. It enables proactive surveillance, personalised treatment strategies, and rapid responses to emerging threats, such as DR, including the identification of critical emerging Plasmodium mutations. This approach not only improves clinical care but also strengthens public health systems by facilitating informed decision-making and promoting collaborative data sharing among researchers and healthcare providers globally. The project will also involve key stakeholders, including Brazil's NMCP, to enhance capacity in AI and genomics through workshops and the development of dashboards and end-user reports. These resources will aid in implementing and validating the informatics platform, incorporating AI functionalities such as spatial analysis for public health applications. These efforts aim to ensure the tools' readiness for clinical and surveillance purposes, thereby contributing to reducing malaria and other infectious diseases in Brazil and aligning with the World Health Organization's regional elimination goals and global health objectives.