Prediction of interactions between pairs and triplets of genes in Saccharomyces cerevisiae using Mixed-Membership Stochastic Block Models

Identifier:  TFG:3033

Handle: https://hdl.handle.net/20.500.11797/TFG3033

Authors:  Mariné Tena, Aleix

Abstract:
In this project we implement a mathematical model based in Mixed-Membership Stochastic Block Models to be able to make predictions of the strength and type of genetic interaction between two or three genes from the human model organism Saccharomyces cerevisiae. We use a data-set of 501510 entries obtained from the supplementary materials of the article “Systematic Analysis of Complex Genetic Interactions”. This data-set contains the fitness data from yeast triple and double knock-out mutants, each with a different combination of mutated genes. After validating the predictions of the model using different metrics, we compare how genes are related according to Mixed-Membership Stochastic Block Models are related in Gene Ontology terms.