John P. C. Le Quesne

This study identifies YBX3 as an oncoprotein in mesothelioma that drives cell proliferation through SLC7A5/LAT1-mediated amino acid uptake and shows JPH203 as a potential therapeutic agent.

This paper uses self-supervised learning to discover histomorphological phenotypes from mesothelioma biopsies, linking tissue patterns to molecular and clinical outcomes.

This study identifies mTORC2 as a convergent dependency in lung cancer that stabilizes HIF-1β to drive metabolic adaptation and tumor progression, revealing a therapeutic target in sphingolipid metabolism.

This study identifies Podocalyxin as a regulator of glandular architecture in colorectal cancer, showing its role in promoting tumour growth and metastasis.

The paper identifies high MYC activity as a targetable feature in lung adenocarcinoma using phenotype-based subtyping and shows sensitivity to a dual inhibitor of MYC co-factors.

This study identifies architectural subtypes of PDAC using morphospatial profiling of CAFs, showing improved prognostication and chemotherapy response prediction.

This study uses spatial transcriptomics to analyze gene expression in lung adenocarcinoma morphologies, revealing plasticity and gene expression trajectories linked to lethal growth patterns.

This paper uses self-supervised AI to identify under-appreciated stromal morphologies in lung adenocarcinoma linked to poor prognosis, independent of current clinical schemas.

This paper introduces a self-supervised learning framework to analyze multiplex immunofluorescence images, revealing novel prognostic spatial patterns in cancer tissues.

This paper uses self-supervised learning on 3446 H&E-stained mesothelioma slides to create a histomorphological atlas, achieving high predictive performance and revealing molecular insights.

This study shows that increased mRNA translation delays lung adenocarcinoma initiation and creates a vulnerability to MEK inhibitors in KRAS-driven tumors.

This study shows that activation of the integrated stress response (ISR) drives tumor evolution and therapy resistance by promoting dedifferentiation and stemness in lung adenocarcinoma.

PixlMap is a deep learning method for cellular phenotyping in multiplex immunofluorescence images that uses nuclear segmentation alone to classify cells without requiring full cell boundaries.

The paper shows copper induces p53 unfolding in cancer cells, leading to invasion and chemoresistance, with partial restoration by the chelator Trien.

This study shows eIF4A1 is critical for supporting translational demands in Wnt-driven colorectal cancers, while eIF4A2 loss accelerates tumourigenesis, and eFT226 inhibits eIF4A1 function in these models.

The paper identifies AG-rich RNA sequences in 5'UTRs that regulate eIF4A1 multimerisation to enhance translation of oncogenic mRNAs.