Supercomputer maps spliceosome in 2M-atom simulation
Get the Health newsletter
Daily health & science — research, biotech, public health, the studies worth knowing. Free.
- Italian Institute of Technology researchers, led by Marco De Vivo with first author Ph.D. student Gianfranco Martino, simulated spliceosome dynamics in a ~2 million atom model — roughly 4-10× larger than typical molecular simulations of 200,000–500,000 atoms
- The "Franklin" supercomputer at IIT, powered by more than 360 GPU processors, enabled atomic-level tracking of how the spliceosome changes shape during activity
- The study, a collaboration with Uppsala University and AstraZeneca, was published in Proceedings of the National Academy of Sciences and helps explain experimental splicing data that had previously been difficult to interpret
- Researchers discovered that spliceosome movements follow a precise, controlled sequence — filling a gap left by the static snapshots scientists previously had of the complex
- De Vivo's team plans next to refine molecules already identified as capable of regulating spliceosome activity, with potential applications in cancer and neurodegenerative disease treatments
- Uppsala's Marco Marcia, an RNA experimentalist collaborating with De Vivo's computational drug discovery group, called the cross-disciplinary pairing key to moving faster toward new therapies
Why it matters: Until now, scientists could only freeze the spliceosome in static snapshots; De Vivo's 2-million-atom dynamic model is the first to show the controlled motion sequence that drives the first step of splicing. That mechanistic clarity gives drug developers a concrete target for designing molecules that can modulate splicing — directly relevant to cancer and neurodegenerative diseases where splicing goes wrong.
