Note: This is a preliminary release of CellTx. Please contact the author for access to the latest full release.
Cell Therapy: Cell CAD framework for dynamically constructing and evaluating multiscale models of cell therapies.
celltx is a framework for building, visualizing, and simulating multiscale models of live cell therapies. celltx consists of a series of abstraction layers, which bridge model specifications which are biologically intuitive with model representations that are mathematically rigorous and numerically evaluable. At present, celltx converts a set of biological specifications into a directed graph, and further computes a system of differential equations representing the state of the biological system over time from the graph.
Installing the Package
After cloning this repository, issue the following command from the root directory.
python setup.py install
Or better yet, using pip:
pip install .
Building Documentation
The documentation is managed with Sphinx. To build the documentation, issue the following from the root directory.
cd docs
make html
To regenerate the docs automatically from from the docstrings, use sphinx-apidoc:
cd docs
sphinx-apidoc -o source/ ../celltx/
Biology Layer - tx_cells, cell, cytokines
The biology layer is a lightweight interface that makes it easy to create common systems of compartments, elements, and relationships.
- compartments passthrough to systems layer compartments
- tx_cells (e.g. a synNotch → CAR T cell) are elements which:
- have one or more internal binary states describing circuitry (e.g. primed, activated), which are linked in specific ways using edge functions that depend on compartmental variables
- have birth, death, killing, secretion, and migratory functions that may depend on their states and/or the values of other elements the system.
- tx_cells have circuitry relationships that are always intracompartmental linkages that depend only on compartmental entities.
- cells (e.g. normal or tumor cells) are compartment-specific elements which:
- have birth and death functions that may depend on the values of other elements in the system
- cytokines are elements with a value for each compartment which:
- have degradation functions
- have diffusion functions which govern their migration throughout compartments
- At this level, interactions are not necessarily linkages between states. For instance, target cell killing is a property of Tx cells.
Systems Layer – Compartments, Elements, and Relationships
The systems layer aims to describe the structure of a biological system in a generalized fashion that is still easily interpretable and manipulated.
- Compartments are regions of biological space (e.g. tumor, normal tissue, and circulation) through which species can travel. The compartments in a system have linkages that specify their topology.
- Elements are a generalization of biological species, which can have arbitrary internal states. Elements may be global to the system, or compartment specific.
- Relationships are a generalization of interactions between biological species. Relationships describe the impact of one element state on another using functions which may be dependent on constants and the values of elements in the system.
Graph Layer - Entities and Linkages
The graph layer describes the structure and states of a biological system rigorously in the form of a directed graph.
- Nodes in the graph correspond to entities, which have a scalar magnitude and in general correspond to the amount of an element in a given compartment and state (e.g. the number of primed and unactivated CAR T cells in the tumor)
- Edges in the graph describe the direction and nature of the influence of one entity on another. Only entities that correspond to different states of the same molecule or cell can have linkages. Edges are labelled with edge functions that describe the influence of one entity on another mathematically. These functions may be dependent on constants, helper functions (e.g. hill function), and the historical magnitudes of some entities.
ODE Layer - Differential Equations
The ODE layer describes the behavior of a biological system over time in the form of a number of ordinary differential equations.
- The derivative of the magnitude of an entity is given by a linear combination of the edge functions pointing to that entity node.
- There will be as many equations as there are nodes in the system.
Other Useful Tips
To regenerate the documentation from zero:
cd docs
rm -rf source/*.rst
sphinx-apidoc -o source/ ../celltx/
cp source/celltx.rst source/index.rst
make html