University of British Columbia
Machine learning algorithms in flow cytometry data analysis
Abstract
dc:descriptionThe state-of-the-art approach to identify cell populations in flow cytometry (FCM) data is called "manual gating". The application of manual gating in complex projects that involve a high number of samples and markers is time-consuming, highly subjective and not reproducible. At least 38 bioinformatic tools have been developed in recent years to automate the gating step but they have either low accuracy or a complex setup. I developed the flowMagic algorithm to completely automate the manual gating process providing high accuracy results in a short amount of time. The flowMagic algorithm includes a machine learning model trained on a large FCM dataset gated by the videogamers of "Project Discovery", which is a mini-game within the online game called "EVE Online". The gated data includes a wide variety of immunological data from the public repositories flowRepository, ImmPort and Cytobank. It also includes COVID-19 patients data, newborns immunological data (from the HIPC-EPIC project) and adult immunological data designed for data analysis standardization. The data was processed using the flowSim tool to remove redundant information, improving the quality of the training set used by the flowMagic tool.
Degree
thesis:*- Name thesis:degree_name
- Doctor of Philosophy - PhD
- Level thesis:degree_level
- doctoral
- Discipline thesis:degree_discipline
- Bioinformatics
- Grantor dc:publisher
- University of British Columbia
- Year dc:date
- 2024
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Montante, Sebastiano
- Contributors dc:contributor
-
- Brinkman, Ryan R
Rights
dc:rights- Statement dc:rights
-
- Attribution-NonCommercial-NoDerivatives 4.0 International
- Language dc:language
- eng
Identifiers
dc:identifier.*- Handle dc:identifier
- http://hdl.handle.net/2429/87991