Computational Oncology

Drug failures are
mechanistically visible.

Causal Bio builds ODE-based signaling pathway models to identify why targeted therapies fail — PD-efficacy disconnects, resistance emergence, combination response — before the next trial.

Work with us rohit@causalbio.io
Pathway
PI3K / AKT / mTOR
Pathway
MAPK / ERK / RAS
Pathway
CDK4/6 + RB-E2F
Pathway
BCL2 / MCL1 Apoptosis
How it works
01

Program review

Trial data, biomarker readouts, and clinical context reviewed to identify mechanistically plausible failure modes.

02

Model calibration

Pathway models calibrated against literature-derived constraints spanning multiple biological contexts.

03

Retrospective analysis

Simulations across genotypic and phenotypic contexts identify where biology diverged from the drug's intended mechanism.

04

Prospective prediction

Resistance hypotheses, combination response, and patient stratification with mechanistic justification.

About

Rohit Nandakumar is a computational biologist trained at Carnegie Mellon University. Causal Bio's platform was built from the ground up — pathway topology, ODE architecture, calibration pipeline, and constraint validation — as a unified system for oncology drug response modeling. The approach is deliberately mechanistic: signaling networks modeled as dynamic systems, not static interaction databases.

Training CMU, Computational Biology
Validation 91+ constraints, 17+ contexts
Architecture ODE-based mechanistic modeling
Engagements Retrospective + prospective

Let's look at
your program.

Retrospective analysis on past programs with unclear mechanistic explanations, or prospective prediction for active programs before the next trial decision.

rohit@causalbio.io Send an email