RiboUnmix finds shared ribosome patterns across noisy experiments
RiboUnmix: Learning Shared Translational Dynamics from Biased and Noisy Ribo-seq Measurements
Machine Learning
Summary
Ribosome profiling measures where ribosomes sit on RNA, but the data often includes noise and experiment-specific quirks. The authors developed RiboUnmix, a method that learns common ribosome patterns shared across different noisy experiments by separating true biological signals from technical biases. Testing on synthetic and real datasets shows RiboUnmix improves predictions of ribosome distribution and helps reveal consistent biological patterns. This makes it easier to understand how proteins are made from various experimental data.
What this means in practice
- •For bioinformatics teams: Predict ribosome distribution more accurately by integrating multiple ribo-seq experiments to reduce noise and biases.
- •For pharmaceutical research units: Identify consistent translation patterns useful for understanding drug effects on protein synthesis by analyzing diverse ribosome profiling data.
Authors
Gabriele Martino, Denis Skibinski, Ivo L. Hofacker, Sebastian Tschiatschek
Abstract
Ribosome profiling (Ribo-seq) measures ribosome distributions along mRNAs, but observed occupancy profiles also contain experiment-specific distortions and stochastic variability. Consequently, models that accurately predict measured profiles may reproduce technical effects rather than recover the underlying biology. We ask whether jointly modeling datasets collected under different experimental conditions can reveal shared, sequence-dependent patterns of ribosome occupancy. We introduce RiboUnmix, a probabilistic multi-dataset framework in which each expected measured profile is represented as a shared sequence-dependent signal modulated by a dataset-specific multiplicative factor. A negative-binomial observation model captures variability across replicates. We evaluate RiboUnmix on a controlled synthetic benchmark combining programmed translation kinetics, ribosome traffic, stochastic count sampling, and sequence-dependent experimental distortions. Because the underlying kinetics and distortions are known, recovery of the shared profile and dataset-specific effects can be assessed separately. Both inferred components correlate strongly with their targets, demonstrating that RiboUnmix can disentangle shared kinetic patterns from experimental effects. Across four organism-specific real-data benchmarks, RiboUnmix outperforms sequence-to-profile baselines in predicting measured profiles. Models trained independently on subsets of 114 HEK-derived datasets recover concordant shared profiles for held-out transcripts, and experiments varying the number and composition of training datasets show that the learned representation remains stable. RiboUnmix thus converts variation across experiments into evidence for reproducible sequence-dependent patterns of ribosome occupancy, supporting biological hypothesis generation from diverse Ribo-seq datasets.