Harnessing alignment and annotation-free information from
next generation sequencing experiments to enable tumor
specific antigen identification and aI-powered prediction of 
immune checkpoint inhibitor response

Κανάτα, Ελένη; Kanata, Eleni

dc.contributor.author	Κανάτα, Ελένη	el
dc.contributor.author	Kanata, Eleni	en
dc.date.accessioned	2021-01-04T08:31:39Z
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/52721
dc.identifier.uri	http://dx.doi.org/10.26240/heal.ntua.20419
dc.rights	Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα	*
dc.rights.uri	http://creativecommons.org/licenses/by-nc-nd/3.0/gr/	*
dc.subject	Bioinformatics	en
dc.subject	Immunotherapy	en
dc.subject	Artificial intelligence	en
dc.subject	Cancer	en
dc.title	Harnessing alignment and annotation-free information from next generation sequencing experiments to enable tumor specific antigen identification and aI-powered prediction of immune checkpoint inhibitor response	en
heal.type	bachelorThesis
heal.secondaryTitle	Αξιοποίηση πληροφορίας χωρίς ευθυγράμμιση και χρήση σημειογραφίας από δεδομένα αλληλούχισης νέας γενιάς (NGS), για ταυτοποίηση καρκινικών αντιγόνων και πρόβλεψη της απόκρισης σε αναστολή ανοσολογικού σημείου ελέγχου (ICI) με μοντέλο τεχνητής νοημοσύνης.	el
heal.classification	Bioinformatics	en
heal.dateAvailable	2022-01-03T22:00:00Z
heal.language	en
heal.access	embargo
heal.recordProvider	ntua	el
heal.publicationDate	2020-10-05
heal.abstract	Immunotherapy harnesses the patient's immune system to elicit anticancer response, enabling long-term remission with low side-effects. Inhibition of Immune Checkpoints, central regulators of self-non-self recognition mechanisms prevents cancer from evading the immune system. However, only a fraction of patients responds to immunotherapy. Being able to predict patient response accurately and timely allows for optimal patient stratification, while maximizing therapeutic outcome. Next Generation RNA Sequencing offers valuable high-throughput information on the cancer cells' transcriptome, that could be used to inform immunotherapy decisions, bypassing limitations in current approaches focusing on the patient’s exome. Existing computational tools rely on alignment to a reference genome and its annotation, therefore restricting the information space, and limiting its applicability in this context. This thesis aims to generate beyond the-state-of-the-art RNA-Seq analysis methods and proteogenomic pipelines and apply them to capture a tumor's antigenic profile as well as tο identify personalized immunotherapy targets. It will also generate meaningful predictors that will be incorporated in an artificial intelligence framework and enable for the first time accurate personalized prediction of patient response to Immune Checkpoint Inhibition. The model will be trained and tested on immune checkpoint inhibitor clinical trial data and benchmarked against available markers, scoring systems, and predictors. This system will revolutionize cancer immunotherapy as it will allow quick and reliable clinical decision-making.	en
heal.abstract	Immunotherapy harnesses the patient's immune system to elicit anticancer response, enabling long-term remission with low side-effects. Inhibition of Immune Checkpoints, central regulators of self-non-self recognition mechanisms prevents cancer from evading the immune system. However, only a fraction of patients responds to immunotherapy. Being able to predict patient response accurately and timely allows for optimal patient stratification, while maximizing therapeutic outcome. Next Generation RNA Sequencing offers valuable high-throughput information on the cancer cells' transcriptome, that could be used to inform immunotherapy decisions, bypassing limitations in current approaches focusing on the patient’s exome. Existing computational tools rely on alignment to a reference genome and its annotation, therefore restricting the information space, and limiting its applicability in this context. This thesis aims to generate beyond the-state-of-the-art RNA-Seq analysis methods and proteogenomic pipelines and apply them to capture a tumor's antigenic profile as well as tο identify personalized immunotherapy targets. It will also generate meaningful predictors that will be incorporated in an artificial intelligence framework and enable for the first time accurate personalized prediction of patient response to Immune Checkpoint Inhibition. The model will be trained and tested on immune checkpoint inhibitor clinical trial data and benchmarked against available markers, scoring systems, and predictors. This system will revolutionize cancer immunotherapy as it will allow quick and reliable clinical decision-making.	en
heal.advisorName	Vlachos, Ioannis	en
heal.committeeMemberName	Boudouvis, Andreas	en
heal.academicPublisher	Εθνικό Μετσόβιο Πολυτεχνείο. Σχολή Χημικών Μηχανικών	el
heal.academicPublisherID	ntua
heal.numberOfPages	77 p.	en
heal.fullTextAvailability	false