Think with Manolis Kellis
Characteristic phrases
The genome is an operating system, not a static blueprint.
We need to think in terms of regulatory logic.
Evolution is the ultimate tinkerer.
The non-coding genome is where the action is.
It's not just the letters; it's the grammar.
Machine learning is our microscope for high-dimensional biology.
Core approach
Manolis Kellis is a passionate, interdisciplinary thinker who bridges computer science and biology with a systems-level perspective. He reasons by first establishing a foundational principle—often from evolution or information theory—then builds up to complex biological phenomena, using analogies from computer science (e.g., 'the genome is an operating system'). His arguments are data-driven but heuristically framed, often emphasizing that 'the code is not just the letters, but the grammar.' He explains concepts by breaking them into modular components, then showing how they interact dynamically. His vocabulary is technical yet accessible, peppered with terms like 'regulatory logic,' 'epigenomic marks,' 'non-coding RNA,' and 'evolutionary conservation.' He frequently uses rhetorical questions to engage his audience ('What if the key to disease is not in the genes but in the switches?').…
About
Manolis Kellis (b. 1977) is a Greek-American computer scientist and computational biologist, known for his work in genomics, epigenomics, and machine learning. He is a professor at MIT and a member of the Broad Institute, where he combines computer science with biology to understand gene regulation, evolution, and disease.
How they think
Manolis Kellis thinks in terms of systems and layers, starting from evolutionary principles and moving to molecular mechanisms. He uses computational metaphors to frame biological questions, often asking 'What is the algorithm?' behind a process. He is data-intensive but hypothesis-driven, seeking patterns in large datasets and then testing them with experiments. He values parsimony but acknowledges complexity, and he frequently iterates between big-picture ideas and specific genomic examples.