Nil Sahin Wins PhD Thesis Prize
Computational biologist Nil Sahin is hunting down promising drug molecules with the help of artificial intelligence. After completing her PhD last Fall in the Donnelly Centre for Cellular and Biomolecular Research, she joined the AI company Cyclica where she creates algorithms capable of identifying compounds with the most therapeutic potential.
It had always been Sahin’s goal to apply her graduate training to solving problems in medicine, she said.
During her PhD, she developed a computational pipeline for the detection of cellular defects caused by genetic mutations. Her work bridged high throughput microscopy with data analysis in a way that has already accelerated research into the molecular underpinnings of disease, where clues to new treatments might be hiding.
Now her efforts have being recognized with the Donnelly Centre Research Thesis Prize, reserved for a top graduate student in the Centre whose PhD project crosses scientific disciplines while meeting the highest standards of excellence. The winner is selected annually by a judging panel from a pool of graduates who obtained their PhD in the previous 12 months.
Sahin obtained her PhD in the laboratory of Brenda Andrews, University Professor of molecular genetics and founding director of the Centre. She was also co-supervised by Quaid Morris, AI expert and former professor of molecular genetics and computer science at the Centre, now at the Memorial Sloan Kettering Cancer Centre in New York.
“I would like to offer my warmest congratulations to Nil on this highly deserved award,” said Jason Moffat, Professor of molecular genetics in the Centre and chair of the award panel.
“Nil’s pioneering work in AI-assisted computer vision has opened new avenues for exploring how genetic mutations and other perturbations affect cellular morphology, with wider implications for understanding the molecular mechanisms of disease and treatment development.”
Donnelly Centre investigators Ben Blencowe, Molly Shoichet, Mikko Taipale and Aaron Wheeler were also among the judges.
Sahin is no stranger to receiving honors. She previously won two flagship Donnelly Centre awards for graduate students, the Cecil Yip Doctoral Research Award for 2017, and the Jennifer Dorrington Graduate Research Award for 2020. But the thesis prize holds a special place for her.
“When you’re doing your PhD — when you’re in the thick of it— it’s difficult to know how much impact you’re making,” said Sahin. “It means a lot to me that other scientists found that the tools I developed during my PhD have made significant contributions to science."
When Sahin started her PhD, computer vision was just beginning its inroads into cell biology labs. As image recognition algorithms transformed everyday lives, from banking to cell phones, cell biology researchers were still analyzing microscopy images by eye—and many still do. This is partly due to a lack of researchers skilled in both cellular biology and software development who can design advanced analysis tools for specific research questions.
Sahin, who received undergraduate training in biology and computer science at Sabanci University in Instanbul, Turkey, was ready to fill this knowledge gap.
She joined Andrews’ lab to study how genetic mutations affect cellular morphology, using the single-celled Baker’s yeast, Saccharomyces cerevisiae, as a model system. The Andrews lab is world renowned for its pioneering work in functional genomics that established how a full complement of genes in a genome interact to sustain the life of yeasts. Over the last decade, they also pioneered automated microscopy approaches, which allowed them to collect images of millions of cells that are either healthy or carry genetic mutations. But the data analysis proved to be a bottleneck.
“We look at all of the genes in yeast and we study all of them,” said Sahin. “A lot of data we produce are so large scale that we need machine learning to be able to tease out what’s interesting and what we’re looking for. But machine learning is not a plug and play easy tool. There are many methods available, and you need to find the correct ones and know how to apply them.”
Machine learning is a form of artificial intelligence that recognizes patterns in large datasets.
Sahin said that although her PhD involved a lot of troubleshooting, which was frustrating at times, she is happy that the analysis pipeline she built is helping other researchers, in the Andrews lab and beyond, address their specific questions.
“The greatest joy for me as a computational scientist is to design software that helps researchers obtain knowledge from data in a fast and reliable way,” she said.
She also added that she is grateful to her mentors who provided more than scientific advice.
“In addition to providing guidance on my project, Brenda made me feel like I could always count on her. Being an international student, she made me feel like the lab was my second home.”
Follow us on LinkedIn and Twitter to keep up with Donnelly Centre news.