Mechanistic and data-informed models help us examine how molecular, physiological and clinical processes evolve together.
Longitudinal and network models for diabetes, insulin resistance and non-alcoholic fatty liver disease to understand the complex nature of the metabolic disease progression. We use AL/ML tools to identify features that indicate the progression of these metabolic diseases.
Logical and dynamical representations of interacting antiviral, inflammatory and tissue-injury pathways. We curated virus-immune response network from literature and modeled its regulatory interactions using a Boolean logic framework. Biological pathways were then mapped into logical equations using 'AND' and 'OR' operators to represent conjunctive and alternative immune signals.
We are developing an ODE-based mechanistic model that captures the distinct cellular interactions of T-DXd in HER2-low tumors. By integrating patient-specific data into the model, we aim to predict whether the therapy will be effective for an individual patient. For patients predicted to respond, the model further estimates the optimal drug dosage required to achieve the desired therapeutic outcome.
We translate clinical requirements into simple, measurable and deployable engineering systems.
Specifications begin with the intended user, measurement and operating environment.
Design choices are made understandable, testable and maintainable.
Accuracy, repeatability, latency, reliability and failure recovery are evaluated explicitly.
Cost and practical deployment are treated as engineering constraints from the beginning.
Interested in working at the intersection of mechanistic biology, molecular engineering and quantitative modelling? Explore current opportunities to join our research team.
Explore positions