Hatice U. Osmanbeyoglu

DGAT: A Dual-Graph Attention Network for Inferring Spatial Protein Landscapes from Transcriptomics

DGAT imputes spatial protein expression from transcriptomics-only data, revealing cell states and tissue organization not visible from RNA alone.

STAN, a computational framework for inferring spatially informed transcription factor activity across cellular contexts

STAN infers spatially informed transcription factor activity using spatial transcriptomics data from lymph node, breast cancer, and glioblastoma datasets.

Loss of CFIm activates YAP/TAZ and connects mRNA cleavage and polyadenylation inhibition to BRCAness

The paper shows that loss of CFIm activates YAP/TAZ and links mRNA cleavage and polyadenylation inhibition to BRCAness through DNA repair gene repression.

Role of the NuRD complex and altered proteostasis in cancer cell quiescence

This study identifies NuRD complex downregulation and altered proteostasis in quiescent ovarian cancer cells and shows that targeting these mechanisms can synergistically kill cancer cells.

Cancer-cell derived S100A11 promotes macrophage recruitment in ER+ breast cancer

The paper identifies S100A11 as a factor promoting macrophage recruitment in ER+ breast cancer through single-cell RNA-sequencing analysis and functional experiments.

Spatial landscape of malignant pleural and peritoneal mesothelioma tumor immune microenvironment

This study uses spatial analysis to examine immune cell distribution in mesothelioma tumors and its association with gene expression and patient outcomes.

COVID-19db linkage maps of cell surface proteins and transcription factors in immune cells

The paper develops a database linking cell surface proteins and transcription factors in immune cells using CITE-seq data from COVID-19 patients and healthy controls.

Isolated BAP1 loss in malignant pleural mesothelioma predicts immunogenicity with implications for immunotherapeutic response

The paper identifies BAP1 loss as a predictor of immunogenicity and potential response to PD-1 therapy in malignant pleural mesothelioma.

Interpretable deep learning for chromatin-informed inference of transcriptional programs driven by somatic alterations across cancers

The paper introduces CITRUS, a deep learning model that infers transcriptional programs influenced by somatic alterations across 17 cancer types using genomic, transcriptomic, and epigenomic data from The Cancer Genome Atlas.

SPaRTAN, a computational framework for linking cell-surface receptors to transcriptional regulators

SPaRTAN links cell-surface receptors to transcription factors using CITE-seq data and cis-regulatory information, revealing novel receptor-TF relationships in immune and tumor cells.

Chromatin-informed inference of transcriptional programs in gynecologic and basal breast cancers

The paper develops PSIONIC, a machine learning method that integrates chromatin accessibility and RNA-seq data to infer transcriptional programs in gynecologic and basal breast cancers, revealing regulatory differences and survival associations.