School of Mathematical Sciences

A transcriptional analysis of image-localized high-grade glioma biopsies reveals a meaningful graph of tumor development

Date(s)
Tuesday 23rd (14:00) - Thursday 25th April 2024 (15:00)
Description
Speaker's Name: Dr Lee Curtin
Speaker's Affiliation: Mayo Clinic, USA
Speaker's Research Theme(s): CMMB,
Abstract:
High-grade glioma (HGG) portends dismal survival, owing in part to its intra- and inter-patient heterogeneity. A limited number of diagnostic biopsies are taken as part of standard clinical care, but these fail to capture the diversity across tumor regions, including variability in immune and normal cell abundances that play key roles in disease development. Anatomical imaging (i.e., MRI) provides a view of the overall tumor, but limited insight into cell-level variation by itself. By collecting and analyzing image-localized biopsies, we can better understand subpopulation ecologies and interactions that may then be exploited for future therapeutic benefit. We collected 202 image-localized multi-regional biopsies from 58 patients to characterize HGG heterogeneity. Samples were run through Monocle, a reverse graph embedding algorithm that groups samples into states and orders them along developmental trajectories. Gene-set enrichment analyses were implemented to determine their distinct pathway expression. CIBERSORTx, a deconvolution algorithm, was used to predict relative abundances of 7 normal, 6 glioma, and 5 immune subpopulations for each sample. CIBERSORTx-predicted abundances were then overlaid on the Monocle graph, both of which were investigated with respect to imaging-defined regions, treatment status, and patient metadata. Monocle classified HGG into four main states along a three-pronged trajectory, which connects with key imaging regions of the disease. The states showed distinct CIBERSORTx-population ecologies and gene-set expression. These algorithms and image-localized biopsies reveal a low-dimensional trajectory that helps us understand the development and evolution of HGG. This approach may be useful as a basis for the development and calibration of future mathematical models.

Venue: Mathematical Sciences A17
Online Conference Link: https://teams.microsoft.com/l/meetup-join/19%3ameeting_NDQ5NDM1MWUtOTJjYi00YmU4LWE3N2UtN2NmMjEyZDljMWY4%40thread.v2/0?context=%7b%22Tid%22%3a%2267bda7ee-fd80-41ef-ac91-358418290a1e%22%2c%22Oid%22%3a%2262d50ba5-440b-4e67-8547-b6518349911a%22%7d

School of Mathematical Sciences

The University of Nottingham
University Park
Nottingham, NG7 2RD

For all enquiries please visit:
www.nottingham.ac.uk/enquire