Computational RNA Biology

In Simple Terms

Modern sequencing machines produce enormous, noisy streams of data about RNA. Hidden inside are the real signals we care about: where a message starts and ends, whether it carries a chemical mark, how much of it there is, and how it is being read. Pulling those signals out reliably is a mathematical and computational problem as much as a biological one. We build the statistics, algorithms and machine-learning models that turn raw sequencing output into trustworthy measurements, and increasingly we let AI drive the analysis so that biologists can ask questions of their data in plain terms.

The Science

This programme develops the analytical backbone shared across our experimental work. It includes change-point and segmentation methods (Segmentor3IsBack) for annotating transcript boundaries from sequencing coverage, compositional statistics (FracFixR) that recover absolute proportions from fractionated RNA-seq, and stochastic modelling (STE) of translation efficiency that accounts for cell-to-cell variability. We also build deep-learning classifiers for base and modification calling on nanopore signal, and interactive tooling such as VX, a desktop genome and transcriptome browser whose viewing and analysis actions are exposed to AI agents through a built-in interface. The aim is rigorous, reproducible inference that scales from single molecules to whole transcriptomes.

Team

• Dr Nikolay ShirokikhLead
• Prof Alice CleynenLead
• Orlane RossiniPhD student
• Shafi MahmudStaff
• Mr Abel SillyPhD student
• Amélie VernayPhD student
• Thomas MinottoStaff
• Jade GoveasStaff
• Xin HoMasters student
• Ben ChungMasters student
• Faiza ChowdhuryGuest Scientist
• A/Prof Sophie LebreGuest Scientist
• Pratosh SathishkumarGuest Scientist
• Mr Mika Martin NieuwenhuyzenGuest Scientist

Tools & Methods

Key Publications

All publications