Profile banner
Çerağ Oğuztüzün

Çerağ Oğuztüzün

PhD candidate in Computer Science applying AI to biology and medicine

Cleveland, Ohio

Summary

Works at the intersection of AI and biology, focusing on translating machine learning and knowledge-graph methods to precision medicine problems such as patient-level drug repurposing. neurograph+2
Builds open tools and visualizers for genomic discovery (e.g., Tokenvizz) that combine language-model attention and graph representations to improve interpretability and downstream biological analysis. github+2
Has a strong record of empirical, peer-reviewed work spanning bioinformatics and clinical informatics (Bioinformatics, JBI, AMIA), with demonstrated emphasis on interpretable methods and real-world biomedical datasets. oup+2
Bridges academia and industry through internships and collaborations (Foundation Medicine, Janssen/Johnson & Johnson, AMD Research, EDDC), applying ML research in production-relevant biomedical settings. neurograph+1

Work

Education

Projects

Writing

Precision Drug Repurposing (PDR): Patient-level modeling and prediction combining foundational knowledge graph with biobank data

January 1, 2025

Introduces a Precision Drug Repurposing framework that integrates individual-level biobank data (PRS, biomarker expressions, medical history) with a foundational biomedical knowledge graph to enable patient-level drug repurposing; evaluated using Alzheimer's disease case studies and UK Biobank data.

Favicon imagepubmed.ncbi.nlm.nih.gov

Tokenvizz: GraphRAG-Inspired Tokenization Tool for Genomic Data Discovery and Visualization (preprint)

December 1, 2024

Preprint describing Tokenvizz, a graph-based tokenizer and visualization tool for genomic sequences that uses attention scores to construct graphs and supports interactive exploration of sequence relationships.

Favicon imagebiorxiv.org

Leveraging Disease-Specific Topologies and Counterfactual Relationships in Knowledge Graphs for Inductive Reasoning in Drug Repurposing

January 1, 2024

NeurIPS AIDrugX workshop spotlight paper presenting KGɪA, a disease-aware graph augmentation method that uses counterfactual relationships and disease-specific topologies to improve inductive reasoning for drug repurposing.

Favicon imageneurips.cc

Characterizing Disparities in the Treatment of Intimate Partner Violence

January 1, 2023

AMIA Joint Summits proceedings paper presenting a computational and statistical meta-regression framework to characterize demographic and social disparities in IPV treatment outcomes.

Favicon imagepubmed.ncbi.nlm.nih.gov

MotifGenie: a Python application for searching transcription factor binding sequences using ChIP-Seq datasets

January 1, 2021

Bioinformatics application paper describing MotifGenie, a tool for integrative motif analysis across multiple ChIP-Seq experiments.

Favicon imageacademic.oup.com