Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/132526
Title: Predicción del valor de concentración letal media, LC50 y del nivel de toxicidad de compuestos orgánicos para Daphnia Magna usando algoritmos de aprendizaje automático supervisado
Author: Pinos Vélez, Verónica Patricia
Tutor: Rebrij, Romina  
Others: Perez-Navarro, Antoni  
Abstract: Acute toxicity tests to determine the median lethal concentration (LC50) in Daphnia Magna are widely applied to determine the level of aquatic toxicity of different compounds. An alternative to these assays is quantitative structure-activity relationship studies (QSAR). This work proposes the creation of a web application that implemented the best machine learning model of the training of different algorithms both for the prediction of the LC50 value and for the classification by level of toxicity of organic molecules through four molecular descriptors: topological polar surface which considers N, O, P and S (TPSA.Tot), number of hydrogen atoms attached to heteroatoms (H.050), octanol-water partition coefficient calculated from the Moriguchi model (MLOGP) and the topological index which encodes information on molecular size and branching, without considering heteroatoms (RDCHI). The algorithms used to obtain the regression models were cubist regression tree, support vector machines with radial kernel (SVMr), random forest (RF), random forest of ranger type, and stochastic gradient impulse (gmb). To generate the classification models, the following will be used: SVM radial, RF, ranger, gmb and neural networks. In both cases, the best model was obtained using SVM with radial kernel. The LC50 prediction model reached a Q2 of 0.77 and an R2 of 0.83 in the external validation and the classification model reached a precision of 0.80.
Keywords: acute toxicity
Daphnia Magna
machine learning
LC50
Document type: info:eu-repo/semantics/masterThesis
Issue Date: 28-Jun-2021
Publication license: http://creativecommons.org/licenses/by-nc-nd/3.0/es/  
Appears in Collections:Trabajos finales de carrera, trabajos de investigación, etc.

Files in This Item:
File Description SizeFormat 
vpinosvTFM0621memoria.pdfMemoria del TFM5,44 MBAdobe PDFThumbnail
View/Open