Analysing Genetic Variation in Ebolaviruses and Cancer Cell Lines

With the arrival of the -omics era and the democratisation of genome sequencing the amount of genetic data is escalating in magnitude orders every year. However, despite all this raw data, the effect prediction of genetic variations in disease remains an open question. The future machine learning algorithms which could solve the problem still require lots of information to feed their development, and it is our mission as bioinformaticians to extract it from the oceans of data.

This Thesis focusses in the analysis of genetic variation in two complete different diseases: Ebolavirus and neuroblastoma.

After the last Ebolavirus outbreak in West Africa (2014), the deadliest one in history, researchers sequenced lots of viral genomes for both surveillance and study of the pathogenic strain. There are still lots to learn from this virus and this Thesis wants to contribute with the study of how it becomes human pathogenic. By comparing different Ebolavirus species, four pathogenic to humans and one not, and looking into functionally important residues called Specificity Determining Positions (SDPs) in their genomes, we predict protein residues which may be key to the host-specifity pathogenicity.

Neuroblastoma is one of the most common cancers in infancy, and the high-risk variety remains a challenging and deadly disease. Chemotherapy is a key treatment for this cancer, so diagnostic of the right drug and effective monitoring of drug resistance emergence could increase the cure ratio of patients. In order to learn more about the genetic variance of this cancer in response to treatment and the effect of these variants in drug resistance emergence, we study the genome of the neuroblastoma cell line UKF-NB-3 and its clonal sub-lines.