Scientific capability / Resistance Min-Cut and Protein–RNA Co-Targeting

Find the path that could keep the signal alive.

A single intervention can leave another route through a signaling network. BioTwin’s resistance model compares protein and RNA co-targeting against the escape paths encoded in the model.

Current work Computational network model

Test the combination against the network

An individual-target ranking tells you which nodes look attractive. In the KRAS-related model, interventions and adaptation change a flow network; the remaining paths show why a proposed combination may still leave signaling capacity. The result is a specific bypass hypothesis to test.

The graph is an explicit hypothesis

Its capacities are hand-set ordinal values, not measured biological rates. The analysis cannot predict patient response from a network diagram. It is most useful when the encoded route can be checked against perturbation evidence in the relevant cell model.

Protein and transcript are different intervention points

Co-targeting may act on protein activity, transcript production or both. Isoform context can change which of those choices makes biological sense. A proposed combined workflow must preserve that context rather than treating every gene product as equivalent.

Test the escape, then the pair

Use single and combined perturbations to ask whether the nominated bypass controls the response. That experiment can support the model, expose a missing route or remove the rationale for the combination.

Make the escape hypothesis testable.

Start with the intervention, the residual signal and the evidence for a compensatory route.