Mehmet Koyutürk

The paper introduces scProfiterole, a framework for clustering single-cell proteomic data using graph contrastive learning with spectral filters, showing improved cell type identification.

This study challenges the ER{beta} dominance hypothesis by showing no significant ER{beta} dominance in endometriosis tissues and highlighting cell- and tissue-specific roles of estrogen receptor isoforms.

This study analyzes phosphoproteome dysregulation in 5XFAD mice, identifying sex-specific and temporal patterns, novel biomarkers, and a key kinase pathway involved in Alzheimer's disease progression.

RokaiXplorer provides an intuitive web tool for phospho-proteomic data analysis and visualization, offering multiple functionalities and interactive features for researchers without specialized training.

The paper evaluates link prediction algorithms in network biology and proposes methods to reduce bias toward well-studied proteins.

The paper develops NetKSA, a network-based machine learning framework to predict kinase-substrate associations, outperforming existing methods and enabling annotation of phosphosites targeted by less-studied kinases.

This paper introduces Macarons, a fast algorithm for selecting complementary genetic variants to predict quantitative phenotypes, achieving similar or better performance than existing methods with significantly reduced runtime.

This paper introduces a method to embed biological networks by considering topological similarity across layers, improving drug repositioning and target prediction accuracy.

The paper investigates co-phosphorylation across diseases, showing functionally related sites exhibit significant co-phosphorylation and can predict pathway or kinase associations.

This paper introduces RoKAI, a network-based framework that improves kinase activity inference by integrating functional information, showing enhanced accuracy and robustness in identifying dysregulated kinases in diseases.

The paper evaluates node similarity measures as convolution matrices in GCN-based methods for link prediction in biomedical networks, showing significant improvements in performance.

The paper develops a network-based algorithm to subtype breast cancer using phosphorylation data, identifying subnetworks correlated with clinical subtypes and reproducible across datasets.

DeepKinZero predicts kinase-phosphosite associations for understudied kinases using zero-shot learning, improving accuracy compared to existing methods.

CoPhosK predicts kinase-substrate associations using co-phosphorylation analysis from mass spectrometry data, improving prediction performance and expanding the known landscape of kinase-substrate interactions.