Aylin is a highly trained mathematician and computational biologist who has published research papers in the fields of mathematics, statistics, health informatics, bioinformatics, machine learning, and biology [1,2,3]. Her research expertise spans several broad and diverse areas including software development, combinatorics, graph theory, genomics, deep learning, and health informatics [1, 4].
Aylin is currently a senior AI scientist at Benevolent AI, a London-based company that uses AI and machine learning for drug discovery, drug repurposing, and precision medicine [1]. Aylin first helped develop a data-driven platform for using unsupervised clustering of electronic health records for precision medicine. She now leads a multidisciplinary team of computational and biology experts that works to evaluate models to enable decision making by precision medicine scientists and senior leadership at Benevolvent AI [1]. Her ambition is to develop ways to better understand the effects of genetic variation on disease and use that data to improve the lives of patients [1]. Aylin has also done an amazing job in the last few years organizing events and organizations promoting the integration of AI, machine learning, and biomedical research. She is the founder of the Crick AI Club at the Francis Crick Institute, which hosts regular seminars and AI events, and which became a local hub for AI researchers in industry and academia [1, 3]. In 2019, she was the primary organizer, fundraiser, and host of the London PhD Network Symposium on Artificial Intelligence, which had over 300 participants and featured speakers from Google, Benevolent AI, UCL and Imperial [1]. In 2018, she was the founder and main organizer of the Crick Data Challenge, an event that brought together computational and experimental scientists to work on research problems involving data analysis of experimental work in a hackathon-style challenge over 2 days [1,3]. This effort led to her winning the Crick 2019 award for “Contributing to or supporting World Class Science” as part of the organising team [1, 3].
Link: Q&A with Aylin Cakiroglu
Research and Educational Background
Aylin did her undergraduate degree in mathematics, with a minor in computer science, at the Technical University Braunschweig, Germany [2, 5]. She began her PhD at the same university in 2008 under Dr. Udo Ott, a well-known expert in discrete mathematics [3]. At the time, she also worked part-time at the National Metrology Institute of Gemany as a mathematician [3].
In September 2008, Aylin visited Queen Mary University of London to collaborate with Dr. Peter J. Cameron, as he was a leading expert in combinatorics which was the branch of mathematics that Aylin was studying at the time [2]. Aylin ended up ended up deciding to stay at QMUL to finish her PhD working on combinatorics, the branch of mathematics which involves studying discrete structures such as ‘graphs’ – networks of connected nodes such as computer networks, transportation networks, or social networks [E]. Under the supervision of Prof Cameron and Professor Rosemary Bailey, Aylin completed her PhD thesis on ‘applying graph theory to the design and analysis of experiments’ [2].
Relevant Link: What is graph theory?
After finishing her undergrad and PhD degrees which were highly mathematical and theoretical, Aylin found that she to do work that was less abstract and theoretical, and had more real-world applications [2] At the time, she found she did not want to pursue an academic career in mathematics, but she also felt that she did not want to go into banking or insurance as most of her peers were doing [2, 5, 6].
She decided to switch into computational biology because she found the field very interesting, challenging, and fun [5]. She ended up taking a postdoctoral position in Professor Nick Luscombe’s group at the Francis Crick Institute in London, where she worked on studying how signals encoded in certain regions of the genome affect gene expression [1, 2, 3, 5]. She was particularly interested how certain recurring patterns in DNA, specifically protein binding sites, may affect gene regulation and applied machine learning and deep learning methods to gain insights into gene regulation [2]. While she was doing her postdoc, she also collaborated with researchers from the Farr Institute of Health Informatics at Kings College London as well as London-based medical doctors [3].
After finishing her PhD, Aylin found she found the research work she was doing was still quite theoretical, as a lot of her research work was in yeast [6]. She found she wanted to do something that would have more tangible impacts on peoples lives, and decided she wanted to focus on translational research work or translating biological and computational findings into real therapeutic strategies [5, 6]. She therefore ended up applying for a position at Benevolent AI, with the hope of doing translational research work to better understand disease and develop pipelines to discover potential therapeutics [6]. She has since led a multidisciplinary team made up of computational and biology experts in developing pipelines for clustering electronic health records and evaluating models for precision medicine [6]
Aylin is passionate about biomedical research, translational work, and AI. She has been a speaker on several discussion panels about AI, machine learning, biomedical research, and women in STEM. She has worked hard to bridge the gap between computational and experimental labs and increase the collaboration between different kinds of scientists to achieve better research goals.
Aylin: "I think the best asset for a woman in bioinformatics are the other women. There is a certain camaraderie amongst us, we all have stories and experiences that reflect our individual set of intersecting minority backgrounds, whether that is gender, sexual orientation, or ethnicity."
Publications:
Aylin has published papers in several different fields, including mathematics, statistics, genomics, and biology. A full list of her publications can be found here. The following are some papers relevant to work mentioned in this article: 1. Cakiroglu A, Clapier CR, Ehrensberger AH, Darbo E, Cairns BR, Luscombe NM, Svejstrup JQ. Genome-wide reconstitution of chromatin transactions reveals that RSC preferentially disrupts H2AZ-containing nucleosomes. Genome Res. 2019 Jun;29(6):988-998. Available at: https://pubmed.ncbi.nlm.nih.gov/31097474/
2. Puthucheary, Z.A., Gensichen, J.S., Cakiroglu, A.S. et al. Implications for post critical illness trial design: sub-phenotyping trajectories of functional recovery among sepsis survivors. Crit Care 24, 577 (2020). Available at: https://ccforum.biomedcentral.com/articles/10.1186/s13054-020-03275-w.
3. S. Aylin Cakiroglu, Judith B. Zaugg, Nicholas M. Luscombe, Backmasking in the yeast genome: encoding overlapping information for protein-coding and RNA degradation, Nucleic Acids Research, Volume 44, Issue 17 (2016). Available at: https://academic.oup.com/nar/article/44/17/8065/2468049?login=true.
Resources
1. Aylin Cakiroglu. Curriculum Vitae. 2021
2. MRC.UKRI. 2018. Postdoctoral research Fellow: Sera Aylin Cakiroglu. [online] Available at: https://mrc.ukDri.org/skills-careers/career-profiles/postdoctoral-research-fellow-sera-aylin-cakiroglu/.
3. LinkedIn. 2021. Aylin Cakiroglu. [online] Available at: https://www.linkedin.com/in/sera-aylin-cakiroglu-2a442973/?originalSubdomain=uk.
4. Lifescihack.com. 2021. LifeSciHack - Conference. [online] Available at: https://www.lifescihack.com/aylin.html
5. BenevolentAI. 2021. Benevolent Stories: Aylin Cakiroglu, Senior AI Scientist. [online] Available at: https://www.youtube.com/watch?v=UXOsKvTnzKo.
6. BenevolentAI. 2021. Women In Tech Meetup: Working in ML Applied Drug Discovery. [online] Available at: https://www.youtube.com/watch?v=tY98HNunSQQ&list=PLWHez0kRXfZ9jwGg6oOZj1gYqNgmD5pcL.