Biology subjects

Fabrini, G.

Publications and source records attributed to Fabrini, G..

5 recordsLinked to original sources

bioRxiv2026

Deep Mechanistic Models reveal pathway-extrinsic drivers of mammary MAPK signalling heterogeneity

Cells sense and respond to their environment through signalling pathways, and the dynamics of these pathways shape cell fate even within genetically identical populations. Two largely separate computational traditions describe this behaviour: mechanistic differential-equation models and representation-learning methods. Mechanistic models encode pathway topology and kinetics but cannot easily represent variation arising outside the modelled pathway. Representation learning, instead, maps genome-wide measurements onto low-dimensional manifolds but offers no mechanistic account of how the resulting cell states execute their functions. Reconciling these views, explaining signalling heterogeneity in a manner that is at once data-driven and mechanistically interpretable, has remained difficult. Here we introduce deep mechanistic models (DMMs), which couple semi-supervised representation learning to an ordinary-differential-equation model of EGFR/MAPK signalling, trained end-to-end so that the learnt representation and mechanistic parametrisation inform each other. Applying DMMs to multiplexed signalling data from 63 breast cancer cell lines, we show that the models generalise to held-out cell lines and attribute most heterogeneity to pathway-extrinsic factors, namely baseline ERBB2 activation and a Ca{superscript 2}/p38 signalling axis, rather than to variation in core MAPK components. Where the models fail, the discrepancies pinpoint rare signalling-altering mutations and recurrent programmes, including a putative AMPK- BRAF MEK-inhibitor-resistance axis and a cytoskeletal programme. We further find that mechanistic integration of EGFR receptor levels reshapes the learnt representation, rendering a molecular and a systems-level account of the same data equivalent in predictive power. DMMs thus offer a general framework for fusing mechanism with learning, naturally extendable to further modalities such as imaging, and simultaneously turn model failure into a systematic route to discover and evaluate candidate biology.

systems biology↗

bioRxiv2025

Expression of nano-engineered RNA organelles in bacteria

Designing synthetic biomolecular condensates, or membrane-less organelles, offers insights on the functions of their natural counterparts, and is equally valuable for cellular and metabolic engineering. Choosing E. coli for its biotechnological relevance, we deploy RNA nanotechnology to design and express non-natural membrane-less organelles in vivo. The designer condensates assemble co-transcriptionally from branched RNA motifs interacting via base-pairing. Exploiting binding selectivity we express orthogonal, non-mixing condensates, and by embedding a protein-binding aptamer we achieve selective protein recruitment. Condensates can be made to dissolve and re-assemble upon thermal cycling, thereby reversibly releasing and re-capturing protein clients. The synthetic organelles are expressed robustly across the cell population and remain stable despite enzymatic RNA processing. Compared with existing solutions based on peptide building blocks or repetitive RNA sequences, these nanostructured RNA motifs enable algorithmic control over interactions, affinity for clients, and condensate microstructure, opening new directions in synthetic biology and biotechnology.

synthetic biology↗

bioRxiv2023

Modular RNA motifs for orthogonal phase separated compartments

Recent discoveries in biology have highlighted the importance of protein and RNA-based condensates as an alternative to classical membrane-bound organelles for the task of compartmentalizing molecules and biochemical reactions. Here, we demonstrate the rational design of pure RNA condensates from star-shaped RNA motifs. We generate condensates using two different RNA nanostar architectures: multi-stranded nanostars whose binding interactions are programmed via single-stranded overhangs, and single-stranded nanostars whose interactions are programmed via kissing loops. Through rational design of the nanostar interaction sequences, we demonstrate that both architectures can produce orthogonal (distinct and immiscible) condensates, which can be individually tracked via fluorogenic aptamers. We also show that aptamers make it possible to recruit peptides and proteins to the condensates with high specificity. Successful cotranscriptional formation of condensates from single-stranded nanostars suggests that they may be genetically encoded and produced in living cells. We provide a library of orthogonal RNA condensates that can be modularly customized and offer a route toward creating systems of functional artificial organelles.

synthetic biology↗

bioRxiv2023

Co-transcriptional production of programmable RNA condensates and synthetic organelles

Condensation of RNA and proteins is central to cellular functions, and the ability to program it would be valuable in synthetic biology and synthetic cell science. Here we introduce a modular platform for engineering synthetic RNA condensates from tailor-made, branched RNA nanostructures that fold and assemble co-transcriptionally. Up to three orthogonal condensates can form simultaneously and selectively accumulate guest molecules. The RNA condensates can be expressed within synthetic cells to produce membrane-less organelles with controlled number, size, morphology and composition, and that display the ability to selectively capture proteins. The in situ expression of programmable RNA condensates could underpin spatial organisation of functionalities in both biological and synthetic cells.

synthetic biology↗

bioRxiv2023

Sculpting DNA-based synthetic cells through phase separation and phase-targeted activity

Synthetic cells, like their biological counterparts, require internal compartments with distinct chemical and physical properties where different functionalities can be localised. Inspired by membrane-less compartmentalisation in biological cells, here we demonstrate how micro-phase separation can be used to engineer heterogeneous cell-like architectures with programmable morphology and compartment-targeted activity. The synthetic cells selfassemble from amphiphilic DNA nanostructures, producing core-shell condensates due to size-induced de-mixing. Lipid deposition and phase-selective etching are then used to generate a porous pseudo-membrane, a cytoplasm analogue, and membrane-less organelles. The synthetic cells can sustain RNA synthesis via in vitro transcription, leading to cytoplasm and pseudo-membrane expansion caused by an accumulation of the transcript. Our approach exemplifies how architectural and functional complexity can emerge from a limited number of distinct building blocks, if molecular-scale programmability, emergent biophysical phenomena, and biochemical activity are coupled to mimic those observed in live cells.

biophysics↗