Biostatistics & High-Dimensional ML Scientist

G ul Bahar B ulb ul

*******@****.*** — gbulbul.com — linkedin.com/in/gbulbul

Professional Summary

PhD-trained statistician with expertise in biostatistics, machine learning, and high-dimensional data analysis, with applications in biomedical and healthcare research. Experienced in predictive modeling, feature engineering, clinical text analysis, and biological network modeling. Strong background in developing statistical and machine learning methods for complex structured datasets, including RNA structural data, cancer-related biological networks, and healthcare-related datasets. Interested in big data science in healthcare, biomedical informatics, and interdisciplinary public health research.

Research Interests

• Big data analytics in healthcare and public health

• Statistical/machine learning methods for healthcare applications

• Predictive modeling and feature engineering

• Biostatistics and causal inference

Education

PhD in Statistics 2023

Bowling Green State University

Advisor: Prof. Craig L. Zirbel

Dissertation: Predicting Base Conservation Scores in RNA 3D Structures (PDF) MSc in Statistics 2019

Middle East Technical University

Advisor: Prof. Vilda Purutcuoglu

Thesis: Novel Model Selection Criteria on High-Dimensional Biological Networks with Applications to Gynecological Cancer Gene Networks (PDF)

BSc in Statistics 2017

Middle East Technical University

Graduated 2nd in class

1

Research Experience

Independent Research 2024–Present

(parental leave since 2024)

• Developing statistical models for RNA conservation analysis and structured biological data

• Investigating dependence structures in sequence and high-dimensional data

• Exploring feature engineering strategies to improve prediction performance

• Preparing research findings for manuscript submission PhD Research and Teaching Assistant, Bowling Green State University 2019–2023

• Conducted statistical analysis on RNA 3D structural datasets using Python

• Developed neural network models to predict base conservation scores

• Identified important predictors influencing biological conservation patterns

• Worked with high-dimensional biological data and computational modeling workflows

• Contributed to interdisciplinary research at the intersection of statistics, machine learning, and biological data analysis

Master’s Thesis Research, Middle East Technical University 2017–2019

• Applied graphical models to high-dimensional biological datasets

• Developed machine learning approaches for gene network construction

• Identified gene sets associated with gynecological cancers

• Performed statistical analysis and model development in R Selected Projects

Statistical and Machine Learning Methods for Biomedical Data

• Applied causal machine learning and feature selection methods to breast cancer subtype analysis

(BRCA dataset)

• Used PCA and statistical modeling to identify meaningful patterns in high-dimensional cancer data

Dementia Detection Using Deep Learning

• Developed neural network models for classification and prediction tasks in a healthcare setting NLP Analysis of Clinical Notes

• Performed text analysis on physician notes for healthcare-related data exploration Skin Cancer Classification

• Developed machine learning models for medical image classification Healthcare Policy Analysis and Predictive Modeling

• Conducted data-driven healthcare policy analysis and predictive modeling using real-world data, focusing on the relationship between insurance coverage and health outcomes 2

Publications

Journal Articles

• B ulb ul, G. B., and Purutcuoglu, V. (2019). Novel Model Selection Criteria on Simulated Sparse Biological Networks. International Journal of Environmental Science and Technology, Springer.

• B ulb ul, G. B., and Purutcuoglu, V. (2021). Novel Model Selection Criteria for LMARS. Journal of Statistical Computation and Simulation, Taylor & Francis.

• B ulb ul, G. B., and Zirbel, C. How RNA Base Conservation Depends on the 3D Context. Manuscript in progress.

Conference Proceedings

• B ulb ul, G. B., and Purutcuoglu, V. (2018). Information Complexity Criterion in the Gaus- sian Graphical Model: Real Data Applications, Proceeding of the International Conference on Mathematics (ICOMATH 2018), Istanbul, Turkiye.

• B ulb ul, G. B., and Purutcuoglu, V. (2018). Model selection in MARS-constructed biological networks, Proceeding of the 5th International Conference on Computational and Experimental Science and Engineering (ICCESEN 2018), Antalya, Turkiye. Selected Conference Presentations

• Entropy-Based Graph Approach in Gene Clustering, JSM 2021 (Online)

• Loop-Based MARS Models, JSM 2020, Philadelphia, USA

• LMARS with STARS Criterion, AMMCS 2019, Istanbul, Turkey

• Inference in Sparse MARS Models, ICDATA 2018, Yalova, Turkey Teaching Experience

Teaching Assistant 2019–2023

• Taught undergraduate-level Introduction to Statistics courses

• Led recitation sessions and held weekly office hours

• Graded assignments and exams and provided detailed feedback to students Technical Skills

Statistical Modeling: Linear regression, logistic regression, survival analysis, Cox models, causal inference, high-dimensional data analysis

Machine Learning: Random Forest, Gradient Boosting, MARS Deep Learning: Neural Networks

Programming: Python, R, SQL

Tools: TensorFlow, PyTorch, Keras

Feature Selection and Engineering: Random Forest importance, LASSO, SVM, clustering methods

3

Contact this candidate