Palak G. Patel

Palak G. Patel

Auto-generated profile, based on publicly available information

Palak G. Patel

Department of Pathology and Molecular Medicine Queen's University Kingston Ontario Canada

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
21 August 2019 in The Prostate
Verified
Journal Article
Closed access
PDF file
PPPalak G. Patel
Palak G. Patel
TWThomas Wessel
Thomas Wessel
AKAtsunari Kawashima
Atsunari Kawashima
JOJohn B. A. Okello
John B. A. Okello
TJTamara Jamaspishvili
Tamara Jamaspishvili
DBDavid M. Berman
David M. Berman

17 June 2025 in bioRxiv
Verified
Preprint
Open access
SPSushant Patkar
Sushant Patkar
TRTimothy R. Rosean
Timothy R. Rosean
PPPalak Patel
Palak Patel
SHStephanie Harmon
Stephanie Harmon
PCPeter L. Choyke
Peter L. Choyke
BTBaris Turkbey
Baris Turkbey
The tumor microenvironment (TME) is a critical focus for biomarker discovery and therapeutic targeting in cancer. However, widespread clinical adoption of TME profiling is hindered by the high cost and technical complexity of current platforms such as spatial transcriptomics and proteomics. Artificial Intelligence (AI)-based analysis of the TME from routine Hematoxylin & Eosin (H&E)-stained pathology slides presents a promising alternative. Yet, most existing deep learning approaches depend on extensive high-quality single-cell or patch-level annotations, which are labor-intensive and costly to generate. To address these limitations, we previously introduced HistoTME, a weakly supervised deep learning framework that predicts the activity of cell type-specific transcriptomic signatures directly from whole slide H&E images of non-small cell lung cancer. This enables rapid, high throughput analysis of the TME composition from whole slide H&E images (WSI) without the need for segmenting and classifying individual cells. In this work, we present HistoTME-v2, a pan-cancer extension of HistoTME, applied across 25 solid tumor types, substantially broadening the scope of prior efforts. HistoTME-v2 demonstrates high accuracy for predicting cell type-specific transcriptomic signature activity from H&E images, achieving a median Pearson correlation of 0.61 with ground truth measurements in internal cross- validation on The Cancer Genome Atlas (TCGA), encompassing 7,586 WSIs, 6,901 patients, and 24 cancer types, and a median Pearson correlation of 0.53 on external validation datasets spanning 5,657 WSIs, 1,775 patients and 9 cancer types. Furthermore, HistoTME- v2 resolves the spatial distribution of key immune and stromal cell types, exhibiting strong spatial concordance with single-cell measurements derived from multiplex imaging (CODEX, IHC) as well as Visium spatial transcriptomics, spanning 259 WSI, 154 patients, and 7 cancer types. Overall, across both bulk and spatial settings, HistoTME-v2 significantly outperforms baselines, positioning it as a robust, interpretable and cost-efficient tool for TME profiling and advancing the integration of spatial biology into routine pathology workflows.

19 September 2026 in The Prostate
Verified
Journal Article
Open access
PDF file
YXYuandong Xing
Yuandong Xing
MXMinqi Xu
Minqi Xu
DDDan Dion
Dan Dion
VBVanessa F. Bratti
Vanessa F. Bratti
ANAnia A. Namin
Ania A. Namin
DBDavid M. Berman
David M. Berman
Background: For prostate cancer (PCa), early diagnosis and biomarker‐based patient stratification can improve clinical decision‐making and post‐treatment prognosis. PCa biomarkers are best validated in tissue following comprehensive histopathologic review; however, urinary biomarkers can be obtained non‐invasively and are therefore safer and more practical to collect and test. It is important to understand how urinary biomarkers can be developed based on gene expression patterns identified in tissue. As a critical step, comparing gene enrichment between prostate tissue and urine may help identify transcripts that are highly enriched in urine and suitable as biomarker candidates. Methods: Using a custom RNA panel enriched for prostate biomarkers, we profiled 367 gene transcripts in prostate tissue ( n = 54) and urine ( n = 93) samples. Based on public protein and expression data for these transcripts, we curated prostate epithelial and stromal gene sets. To evaluate the effectiveness of various data processing methods in making tissue and urine data comparable, we developed an Index of Tissue‐Urine Comparability ( I TUC ). Using the best method identified based on the I TUC to process the data, we analyzed the enrichment of (i) the epithelial and stromal gene sets and (ii) cell marker gene sets. Results: The epithelial gene set ( e.g ., KLK3 ) and stromal gene set ( e.g ., ADAM33 ) comprised 19 and 13 genes respectively. The two gene sets were enriched in their corresponding tissue compartments. To improve the comparability between prostate tissue and urine samples, the ComBat function with parameters “mean.only = T, par. prior = F” proved most effective in the independent training and test datasets. Compared to prostate tissue samples, urine samples showed higher enrichment of the epithelial gene set, as well as the marker gene sets of vascular endothelial cells and regulatory T (T reg ) cells. In contrast, tissue samples were enriched for the marker gene sets of blood fibroblasts and prostate basal cells. Conclusion: Genes enriched in prostate tissue differ from those in urine. Future studies on urine biomarkers could focus on the transcripts derived from prostate epithelial cells, vascular endothelial cells, and immune cells, given their enrichment in urine.