Flash-PFlash-P

Documentation

Complete guide to Flash-P: agents, scripts, configuration, and extension.

Getting Started

Quick Start Guide

Overview and setup instructions

Installation

Dependencies, subscription setup, environment configuration

Pipeline Overview

Architecture and data flow

Agents

CURATOR Agent

Literature mining and edge extraction

BUILDER Agent

Network construction (the brain)

PERTURBATION Agent

Perturbation dataset creation

VALIDATOR Agent

Validation with comprehensive metrics

REFINEMENT Agent

Iterative improvement with biological reasoning

Scripts

flashp_validator.py

Primary algebraic validation

ode_validator.py

ODE with Hill functions

rwr_validator.py

Random Walk with Restart

Reference

Mathematical Foundations

All formulas: algebraic rules, Hill functions, RWR propagation

Network JSON Format

Network output file structure

Validation Results

Validation output files explained

FAQ

Frequently asked questions

Flash-P

AI-powered causal network construction. Where AI meets expert biology.

c.mitsanis@uq.edu.aud.kainer@uq.edu.au

Product

Desktop AppWhy Flash-PVision

Resources

DocumentationAbout

Cite Flash-P

Mitsanis, C. & Kainer, D. (2026). Flash-P: A Multi-Agent System for Constructing Causal Mechanistic Networks.

© 2026 Flash-P