Nutrient consumption and chemotaxis by cells

I am currently trying to model chemotaxis and realistic nutrient consumption by cells. Chemotaxis behaviour works by retrieving the concentration of the bins adjacent to the bin the cell center is in, and then creating a 3d vector which points in the direction of the concentration gradient. Nutrient Consumption is modelled by finding all the bins that is inside the cell, and reducing their contents to (near) zero.

These two behaviours conflict, as reducing the concentration of all bins inside the cell inadvertently removes any concentration gradient at the cell center, so the cell does not move. I have thought of two solutions: 1) nutrient consumption only occurs in the center of the cell, or 2) the concentration gradient is calculated using the boxes immediately outside the cell.

Solution 1 leads to unrealistic nutrient consumption, as the cell does not consume nearly as many nutrients as it should compared to its size. Solution 2 leads to unrealistic chemotaxis as the cell struggles to move diagonally towards nutrients.

Being able to model chemotaxis and nutrient consumption at the same time is an important implementation for simulating microbial communities. Is there potentially a different method for simulating chemotaxis or nutrient consumption which would allow both of these behaviours to co-exist?


Hi @r_bournes, this should be resolvable by changing the timescale of diffusion vs biological behaviour. You can run multiple diffusion time steps before running a biological time step, and iterate this. In this way, the the depleted voxels will fill up due to diffusion from neighbouring voxels. This of course needs several time steps of diffusion to occur, so you should identify a suitable number of diffusion time steps. Please keep us posted on this matter.

FYI, this is relevant: ST07-multi-scale-simulation