Session 2 : Unveiling Clusters in High-dimensional Cancer Data: Predicting Cancer of Unknown Primary

Presenter Information/ Coauthors Information

Padmapriya Swaminathan, Avera Research InstituteFollow

Presentation Type

Invited

Student

No

Track

Genomics

Abstract

Next-generation sequencing (NGS) technologies have become vital in cancer genomics generating high-dimensional data with hidden meaningful relationships. This complexity presents significant challenges for visualization and downstream analyses. Cancer of unknown primary (CUP) refers to cancer that has spread to one or more metastatic sites with the primary cancer site origin that is unknown. To predict the primary cancer site and gain insights, we employed Uniform Manifold Approximation and Projection (UMAP), a non-linear dimensionality reduction technique using normalized gene expression data from 11284 samples representing 26 cancer types from TCGA (The Cancer Genome Atlas). This approach allowed us to construct and visualize clusters of these cancer types on a two-dimensional map, aiding in the prediction of the CUP for a new patient. This prediction, alongside other genomic testing results will help in identifying the primary site of origin in patients with CUP hence contributing to the knowledgebase for clinicians to recommend suitable treatment options.

Start Date

2-7-2025 8:50 AM

End Date

2-7-2025 9:50 AM

This document is currently not available here.

Share

COinS
 
Feb 7th, 8:50 AM Feb 7th, 9:50 AM

Session 2 : Unveiling Clusters in High-dimensional Cancer Data: Predicting Cancer of Unknown Primary

Dakota A & C (Room 250)

Next-generation sequencing (NGS) technologies have become vital in cancer genomics generating high-dimensional data with hidden meaningful relationships. This complexity presents significant challenges for visualization and downstream analyses. Cancer of unknown primary (CUP) refers to cancer that has spread to one or more metastatic sites with the primary cancer site origin that is unknown. To predict the primary cancer site and gain insights, we employed Uniform Manifold Approximation and Projection (UMAP), a non-linear dimensionality reduction technique using normalized gene expression data from 11284 samples representing 26 cancer types from TCGA (The Cancer Genome Atlas). This approach allowed us to construct and visualize clusters of these cancer types on a two-dimensional map, aiding in the prediction of the CUP for a new patient. This prediction, alongside other genomic testing results will help in identifying the primary site of origin in patients with CUP hence contributing to the knowledgebase for clinicians to recommend suitable treatment options.