Dr. Onur ERDOĞAN

PhD Thesis Defense Announcement

Health Informatics Program

"EnSCAN: Ensemble Scoring for Prioritizing Causative Variants Across Multi-Platform GWAS for Late-Onset Alzheimer's Disease"

According to METU Graduate School of Informatics' official announcement, a machine-learning-based ensemble scoring method called "EnSCAN" was developed to prioritize genetic variants associated with late-onset Alzheimer's disease (LOAD) using multi-platform GWAS (genome-wide association study) data. The thesis was completed in February 2025 under the supervision of Assoc. Prof. Yeşim Aydın Son, with Prof. Cem İyigün as co-advisor.

TÜBİTAK's Software Capability Certified in the United States

TÜBİTAK BİLGEM Software Technologies Research Institute (YTE)

In a CMMI appraisal by Carnegie Mellon University's Software Engineering Institute, TÜBİTAK became the first and only public institution to reach Level 5 — the highest level — in software planning, development, and configuration processes, joining roughly 500 organizations worldwide at this level. Dr. Onur Erdoğan contributed to this institutional achievement, having built the enterprise- and project-level measurement-analysis infrastructure as Measurement & Analysis Lead at TÜBİTAK BİLGEM YTE between 2011 and 2020.

EnSCAN: ENsemble Scoring for prioritizing CAusative variaNts across multiplatform GWASs for late-onset alzheimer's disease

Onur Erdogan, Cem Iyigun, Yeşim Aydın Son · BioData Mining, Vol. 18, Article 20 · DOI: 10.1186/s13040-025-00436-x

A machine-learning framework for identifying genetic variants associated with late-onset Alzheimer's disease. Genome-wide association study (GWAS) data from three different cohorts, obtained with different genotyping platforms, were integrated using a novel method called "post-ML ensemble." The EnSCAN algorithm maps variants to their chromosomal positions and prioritizes them via Random Forest validation; the study identified 43 protein-coding variants with high confidence scores, involving genes linked to neurological and metabolic processes. The article is also indexed on PubMed and PMC.

More effective sprint retrospective with statistical analysis

Onur Erdoğan, Muhammed Emre Pekkaya, Halime Gök · Journal of Software: Evolution and Process · DOI: 10.1002/smr.1933

A study on making Scrum teams' sprint retrospective meetings more effective through statistical analysis; it addresses improving sprint planning through more accurate sizing of product backlog items and how historical data can be used to monitor team performance. The academic-publication result of the CMMI Level 5 measurement-analysis experience at TÜBİTAK BİLGEM YTE.