Evolutionary exploration of cell-level morphogenesis in 2D
Membrane-bound dot sizes show proxy atp molecule distribution, the central cell dot show cell cycle progression. Ground coloring shows aminoacids proxy molecule distribution in it (it doesn't diffuse inside cells yet)
- Cell shape is simulated as a balance of internal pressure (based on cell volume) and the membrane's resting length
- Chemistry defines 40 generic molecule types and generates a reaction graph between them. Molecules inside cells are spatially-distributed along the membrane
- Besides those 40, cells contain 3 proxy molecule types:
- atp - acts as a generic energy resource, consumed or produced in chemical reactions
- aminoacids - generic building material, consumed for gene expression
- cycle - upon crossing the threshold, signals the initiation of mitosis. Expressed as a gene
- Cell division axis is selected akin to how the actual mitotic spindle settles into it's orientation [1] - it spins around until the highest concentration of microtubule-binding proteins on the membrane suspend it - though the spinning is replaced by getting a normal to the first component of PCA decomposition of spatial distribution of binding proteins (it's presumed uniform along the membrane right now, so only the cell shape matters)
- Evolution mutates genes by altering activators/repressors/product. Though it's in a very early stage overall
- Cell -> cell and cell <-> environment molecule diffusion and signaling
- Molecule <-> function mapping (e.g. cell adhesion, signaling, diffusion control)
- Gene <-> molecule <-> reaction relations are sketchy
The primary notebook is cell_shape.ipynb
Results are rendered to visdom