Enhancing Spatial Transcriptomics Analysis by Integrating Image-Aware Deep Learning Methods.

Spatial transcriptomics (ST) represents a pivotal advancement in biomedical research, enabling the transcriptional profiling of cells within their morphological context and providing a pivotal tool for understanding spatial heterogeneity in cancer tissues. However, current analytical approaches, akin to single-cell analysis, largely depend on gene expression, underutilizing the rich morphological information inherent in the tissue. We present a novel method integrating spatial transcriptomics and histopathological image data to better capture biologically meaningful patterns in patient data, focusing on aggressive cancer types such as glioblastoma and triple-negative breast cancer. We used a ResNet-based deep learning model to extract key morphological features from high-resolution whole-slide histology images. Spot-level PCA-reduced vectors of both the ResNet-50 analysis of the histological image and the spatial gene expression data were used in Louvain clustering to enable image-aware feature discovery. Assessment of features from image-aware clustering successfully pinpointed key biological features identified by manual histopathology, such as for regions of fibrosis and necrosis, as well as improved edge definition in EGFR-rich areas. Importantly, our combinatorial approach revealed crucial characteristics seen in histopathology that gene-expression-only analysis had missed.Supplemental Material: https://github.com/davcraig75/song_psb2014/blob/main/SupplementaryData.pdf.

Profiling the molecular and clinical landscape of glioblastoma utilizing the Oncology Research Information Exchange Network brain cancer database.

Background: Glioblastoma exhibits aggressive growth and poor outcomes despite treatment, and its marked variability renders therapeutic design and prognostication challenging. The Oncology Research Information Exchange Network (ORIEN) database contains complementary clinical, genomic, and transcriptomic profiling of 206 glioblastoma patients, providing opportunities to identify novel associations between molecular features and clinical outcomes. Methods: Survival analyses were performed using the Logrank test, and clinical features were evaluated using Wilcoxon and chi-squared tests with q-values derived via Benjamini-Hochberg correction. Mutational analyses utilized sample-level enrichments from whole exome sequencing data, and statistical tests were performed using the one-sided Fisher Exact test with Benjamini-Hochberg correction. Transcriptomic analyses utilized a student's t-test with Benjamini-Hochberg correction. Expression fold changes were processed with Ingenuity Pathway Analysis to determine pathway-level alterations between groups. Results: Key findings include an association of MUC17, SYNE1, and TENM1 mutations with prolonged overall survival (OS); decreased OS associated with higher epithelial growth factor receptor (EGFR) mRNA expression, but not with EGFR amplification or mutation; a 14-transcript signature associated with OS > 2 years; and 2 transcripts associated with OS < 1 year. Conclusions: Herein, we report the first clinical, genomic, and transcriptomic analysis of ORIEN glioblastoma cases, incorporating sample reclassification under updated 2021 diagnostic criteria. These findings create multiple avenues for further investigation and reinforce the value of multi-institutional consortia such as ORIEN in deepening our knowledge of intractable diseases such as glioblastoma.