Repository for Thesis Project entitled "Retrieval of functional microbiome information for grapevine using Deep Learning and Biomedical Ontologies"
Student: Madalena Girão
Supervisor: Francisco Couto
Co-supervisor: Ana Margarida Fortes
Grapevine (Vitis vinifera L.) is a globally significant fruit crop, with Portugal being a key player in grape production (OIV, 2017). Traditional viticulture practices have historically supported rich biodiversity within vineyards. However, the intensification of vineyard management, marked by the widespread use of Plant Protection Products, has resulted in detrimental effects on ecosystems, soil quality, and groundwater contamination. To promote sustainable agriculture, preserving biodiversity is imperative. Plant-associated microbiomes play a crucial role in enhancing plant resilience and adaptation to environmental stressors.
This master project aims at enriching existing metagenomics data from field experiments conducted at the ampelographic collection vineyard of Herdade do Esporão (Alentejo) by leveraging metagenomics data and employing text-mining and meta-analysis techniques. In contrast to traditional methods, our method utilizes natural language processing for extracting domain-specific information from texts [https://doi.org/10.1109/JBHI.2022.3173558]. Deep learning and biomedical ontologies are employed to comprehensively capture data related to bacteria, fungi, and relevant parameters, as well as their interrelationships. The extracted information is then summarized and made accessible through knowledge graph visualization tools [https://doi.org/10.3389/fimmu.2017.01656], and continually refined using machine learning methods. These tools enable efficient and systematic retrieval, filtering, classification, exploration, and visualization of metagenomics information, with valuable applications in the wine production sector, contributing to sustainable viticulture practices.
Microdrygrape Project @ FCUL
Bayer Digital Campus Challenge 2023 (Data-Driven Farming)
Sousa, D., Lamurias, A., and Couto, F. M. (2019). A silver standard corpus of human phenotype-gene relations. NAACL HLT 2019 - 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies - Proceedings of the Conference, 1:1487–1492 (https://aclanthology.org/N19-1152/)
Sousa, D. and Couto, F. M. (2023). K-RET: knowledgeable biomedical relation extraction system. Bioinformatics, 39 (https://academic.oup.com/bioinformatics/article/39/4/btad174/7108769)