Gene expression-based classification of urothelial carcinoma for research and clinical applications
(2023) BINP52 20222Degree Projects in Bioinformatics
- Popular Abstract
- Towards personalized treatment: Improved tools for bladder cancer classification
Cancer research has taken a leap forward with the emergence of precision medicine, which aims to tailor treatments to the unique characteristics of each patient’s tumor. IN the case of bladder cancer, a complex and diverse disease, understanding its molecular subtypes is crucial for accurate prognosis and treatment response prediction. This bioinformatics project has developed a method for classifying urothelial carcinoma (UC), the most common type of bladder cancer, based on gene expression profiles.
Cancer classification into different groups or subtypes is typically based on transcriptomic profiling. This technique is used to analyze the levels of... (More) - Towards personalized treatment: Improved tools for bladder cancer classification
Cancer research has taken a leap forward with the emergence of precision medicine, which aims to tailor treatments to the unique characteristics of each patient’s tumor. IN the case of bladder cancer, a complex and diverse disease, understanding its molecular subtypes is crucial for accurate prognosis and treatment response prediction. This bioinformatics project has developed a method for classifying urothelial carcinoma (UC), the most common type of bladder cancer, based on gene expression profiles.
Cancer classification into different groups or subtypes is typically based on transcriptomic profiling. This technique is used to analyze the levels of expression of genes in a given biological sample. Differences in gene expression patterns can help us identify different tumor subtypes. Thanks to transcriptomic profiling, researchers discovered that UC comprises distinct subgroup with different biological features. Characterizing these subtypes has not only opened new possibilities for personalized treatment but has also helped improve our understanding of bladder cancer biology.
One of the challenges in this field is the lack of standardized methodologies. Studies can differ in both biological and technical aspects, and this often leads to discrepancies in the results. To address this issue, we used a Single-Sample-Predictor (SSP) algorithm. Unlike conventional methods that require complex pre-processing steps, the SSP utilizes binary gene-pair rules to classify UC subtypes and can be applied to a sample in isolation. This technique reduces the impact of variations in sample preparation, enabling reliable classification of new samples across different datasets.
Using a comprehensive dataset including multiple studies generated aon different platforms, we developed a rule-based Random Forest classifier that assigns molecular subtypes according to the Lund Taxonomy classification system developed by the Lund bladder cancer group. The classifier was validated against four independent bladder cancer datasets and the predicted subtypes showed significant associations with both clinical outcomes and biological factors.
This research can have important implications facilitating the translation of scientific discoveries into meaningful clinical applications. Accurate subtype prediction can help clinicians can make informed decisions about personalized treatment options for individual patients.
In conclusion, this bioinformatics project represents a step forward in the field of bladder cancer subtyping. We have developed a robust classifier that assigns molecular subtypes of urothelial carcinoma with high accuracy. This advancement not only enhances patient prognostication and treatment response prediction in clinical settings but also paves the way for future advancements in precision medicine.
Master’s Degree Project in Bioinformatics, 60 credits, 2023
Department of Biology, Lund University
Advisor: Pontus Eriksson
Advisors Department: Department of Clinical Science, Division of Oncology, Lund University (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/9135523
- author
- Aramendia Cotillas, Elena
- supervisor
- organization
- course
- BINP52 20222
- year
- 2023
- type
- H2 - Master's Degree (Two Years)
- subject
- language
- English
- id
- 9135523
- date added to LUP
- 2023-08-29 13:54:36
- date last changed
- 2023-08-29 13:54:36
@misc{9135523, author = {{Aramendia Cotillas, Elena}}, language = {{eng}}, note = {{Student Paper}}, title = {{Gene expression-based classification of urothelial carcinoma for research and clinical applications}}, year = {{2023}}, }