I am a data scientist with two years of experience in drug discovery at a growing biotech start-up. During my postdoc, I developed and implemented state-of-the-art machine learning (ML) and deep learning (DL) models. I used these models for ADMET modeling and early drug discovery of small and large molecules, specifically for hit-identification and in low-data conditions. Additionally, I spearheaded the development of the company’s data science experimentation pipeline using Data Version Control (DVC). I primarily use Python for coding. I utilize scikit-learn and PyTorch for classical ML and DL model development; RDKit and Chembl Structure Pipeline for cheminformatic tasks; SQL for data extraction from databases, such as ChEBML; and Git, DVC and Docker for MLOps. I used AWS for some personal projects, specifically RDS, Lambda and EC2 services. My goal is to become a full stack data scientist and one of the top voices in an industry which has a positive impact on human lives.

Academic Background

I received my Ph.D. in Biology from the University of North Carolina at Chapel Hill in December 2022. In my doctorate, I worked with Dr. Laura Miller who is now a professor of mathematics at the University of Arizona. In my doctorate, I investigated the movement of saltwater plankton at different spatial scales using statistical modeling, computational, and experimental fluid dynamics. During my master’s studies, I developed a conceptual underwater vehicle inspired by squid propulsion. I received my masters in Marine Technology from Istanbul Technical University in 2014. Before my passion for biology, I studied Naval Architecture Engineering at Yildiz Technical University and received my bachelor’s degree in 2010.

Why Data Science and Drug Discovery

My academic background is often a conversation piece. It is a great story, and I enjoy talking about it. I would be happy to discuss it with you, as well. Briefly, I began questioning why most man-made marine vehicles utilize propellers. Nature offers a wide variety of solutions, including cilia, rowing propulsion, flapping, and metachronal swimming. Problems and solutions found in natural systems are intellectually much more appealing to me than those in man-made systems. That being said, academia primarily focuses on knowledge generation but less on problem-solving. Perhaps because I got the engineering bug in my undergraduate, I decided to channel my energy into solving real-life problems rather than merely investigating intellectually appealing natural mysteries.

I have started coding in Matlab during my undergraduate. In my master’s, I switched to Fortran for scientific computing. In my Ph.D., I used Matlab and Python in my thesis. Also, during my Ph.D., I began collecting data and analyzing it for finding the behavioral patterns of plankton. It was almost like a detective job. Finding connections between the movement behavior of plankton and their evolutionary past… This experience led me to develop a passion for data science. My transition to drug development occured after I completed an internship at a drug discovery start-up in 2022. It became clear to me after that moment that I wanted to utilize my data science skills to solve real-life problems in an industry which has a positive impact on human lives.