Unit:
Κατεύθυνση Εφαρμοσμένα ΜαθηματικάLibrary of the School of Science
Author:
Choudalakis Stamatios
Supervisors info:
Στρατής Ιωάννης, Καθηγητής, Τμήμα Μαθηματικών, ΕΚΠΑ,
Δικαίος Νικόλαος, Ερευνητής Γ, Κέντρο Ερευνών Θεωρητικών και Εφαρμοσμένων Μαθηματικών, Ακαδημία Αθηνών
Καστής Γεώργιος, ερευνητής Α, Κέντρο Ερευνών Θεωρητικών και Εφαρμοσμένων Μαθηματικών, Ακαδημία Αθηνών
Original Title:
Network Graphs of Cancer Mutations
Translated title:
Network Graphs of Cancer Mutations
Summary:
Cancer is a leading cause of death worldwide, accounting for nearly 10 million
deaths in 2020. Genetic alterations are one of the main causes of the disease’s
initiation and progression, but not all of them confer to it. A significant amount
of effort has been made to distinguish cancer-specific genes, in a sense that mutations
affecting certain genes can be linked with particular cancer types. Here,
a Network Graph built by 8386 patients and a total of 198 genes, from TCGA
data, has been used to find patient-gene groups with the parallel use of two
methodologies separately, one based on the edges between two nodes, using
Barber’s modularity index, and the other focusing in higher-order structures
regarding flow dynamics, using random walks and minimal description statistics.
For each of the clusters recovered, a per-cancer subgroup was taken into
consideration if 10 or more patients of the same type co-existed in the cluster.
Then, the relation between that cancer type and the genes of the cluster was
estimated with respect to current literature. Through this analysis, the union
of the clustering methods revealed known driver genes for all but three cancer
types.
Main subject category:
Science
Keywords:
cancer, graph theory, oncogenes, single point mutations
Number of references:
188
